Professional Documents
Culture Documents
Beyond Borders: Satellite Applications For Humanitarian Emergencies
Beyond Borders: Satellite Applications For Humanitarian Emergencies
ACKNOWLEDGEMENTS
• Ben Ramalingam
• Mark Beagan
• Tonia Thomas
Caribou Space also thanks the project’s Advisory Group (in organisational
alphabetic order) for their valuable contributions to this report.
RECOMMENDED CITATION
www.caribou.space contact@caribou.space
Contents
Executive summary...........................................................................................................3
Background..................................................................................................................... 9
Humanitarian Innovation Hub................................................................................. 10
Satellite applications.................................................................................................11
Humanitarian emergencies.............................................................................................12
What are humanitarian emergencies?.......................................................................... 13
Research framework................................................................................................. 14
Domains of humanitarian emergencies........................................................................ 14
Humanitarian events............................................................................................... 16
Environmental and situational data for humanitarian decision-making...........................19
Ecosystem of satellite applications for humanitarian emergencies................................... 23
Supply-side actors.................................................................................................... 25
Demand-side actors..................................................................................................28
Enabling environment............................................................................................. 31
Satellite applications...................................................................................................... 32
Use cases of greatest need........................................................................................... 33
Use of satellite applications across the event life cycle....................................................... 37
Use cases by domain..................................................................................................40
Satellite applications industry .........................................................................................45
Satellite Earth Observation (EO)............................................................................... 47
Global Navigation Satellite System (GNSS)................................................................. 52
Satellite Communications (SatComms)........................................................................ 54
Landscape of humanitarian satellite applications.............................................................57
Geography.............................................................................................................. 58
User types............................................................................................................... 61
Supplier types......................................................................................................... 61
Domains and events................................................................................................. 63
Business model........................................................................................................ 65
Scale......................................................................................................................66
Data gaps versus availability..................................................................................... 67
Overcoming barriers to uptake of satellite applications in humanitarian emergencies......68
Barriers to uptake.................................................................................................... 69
Priority barriers for development community interventions............................................. 85
Conclusions....................................................................................................................87
Annex 1: Philippines Typhoon Haiyan Case Study......................................................... 90
Annex 2: East Africa Case Study.....................................................................................91
Annex 3: Methodology and limitations...........................................................................93
Annex 4: Glossary.......................................................................................................... 96
Caribou Space
Caribou Space supports organisations to bridge the space and and sustainable
development worlds by working with governments, space agencies, development
agencies, and private sector space companies. Caribou Space provides:
• Fund management: Large-scale ODA funds (£100M+) and seed-stage funds (£4M+).
• Monitoring and Evaluation (M&E): Design of M&E systems, delivery of process and
impact evaluations, and M&E training.
• Economic evaluation: Quantification of the economic cases and impacts of space technology.
2
Executive
summary
Executive summary
Background
The objective of this report is to provide a consolidated view on the current use
of satellite applications in humanitarian settings. Satellite applications are digital
Beyond Borders: Satellite Applications for Humanitarian Emergencies
services and products that serve several functions for society, the environment, and
the economy. UKHIH selected a consortium of organisations to author this report,
including Caribou Space, Satellite Applications Catapult (SAC), and the United
Nations Institute for Training and Research (UNITAR).
Overview
• A summary of two case studies from Asia-Pacific and East Africa to provide real-world
insight from specific regions.
Executive summary
4
Humanitarian emergencies
Humanitarian emergencies are “events or series of events that represent a critical threat
to the health, safety, security, or wellbeing of a community or other large group of people.”
They are widely geographically distributed, often covering large areas and multiple
populations. They are diverse and complex, and require coordinated efforts and
resources across a variety of local, national, and international stakeholders.
• Satellite Earth Observation (EO) is the gathering of information about the physical,
chemical, and biological systems of the planet via remote-sensing technologies.
Platform and cloud services, together with technologies like machine learning, are
simplifying access to and use of satellite applications. In parallel, new entrants and
developments in SatComms and Satellite Internet of Things (IoT) have the potential to
enable new and more cost-effective connectivity to people and things in the near future.
Satellite applications
The most persistent and extensive humanitarian data gaps include a lack of information
on affected schools, malnutrition, damaged infrastructure like buildings and roads, and
refugees, internally displaced persons, and persons of concern. Satellites are valuable
in addressing data gaps across all of these areas for every humanitarian emergency.
Satellite applications support all stages of the event life cycle, from Hazard and Risk
Assessment to Recovery. Satellite EO can be particularly powerful in supporting
predictive analytics for slow-developing crises, for example, in the context of food
insecurity predicting weak crop yields months before the harvest season, or to anticipate
violence and understand the drivers of conflict.
adoption by humanitarian actors across a wide range of use cases. Such use cases
include population and poverty mapping, infrastructure mapping and exposure, crop
yield and productivity forecasting, hazardous event detection and severity evaluation,
and many other applications across humanitarian domains.
5
Environmental and situational data for humanitarian decision-making
The World Bank’s Data for Better Lives report highlights how the production
and collection of data enable humanitarian and development outcomes. Satellite
applications, as digital technology, provide a unique source of such data about people
and their surrounding environment—with the potential to fill significant data and
information gaps in humanitarian management and decision-making processes.
Satellite applications provide:
Data from satellite applications has the specific advantages of coverage, objectivity,
repeatability, thematic detail, speed, and affordability.
side, or both in a few cases. The supply-side provides the satellite applications, whereas the
demand-side uses those applications to address their humanitarian emergencies.
• Private suppliers: satellite operators and resellers, cloud computing providers, platform/
solution providers, and hardware/software suppliers
• Academia
• NGOs
• Media
• Development Agencies
• Governments
• First Responders
Executive summary
• Private Sector
• Academia
• NGOs
• Media
6
• Development Agencies
• A ffected Public
• General Public
Landscape of humanitarian satellite applications
This report includes a landscaping of over 500 humanitarian satellite applications which
identified that:
• The highest number (92) of identified applications are for “food, security, nutrition and
famine” events in Africa.
• Many of the data gaps identified by Humanitarian Data Exchange (HDX) in sectors
such as health and education could be addressed by satellite applications.
• Data availability
• Procurement challenges
7
• Recommended development community interventions are defined for the eight
prioritised barriers.
Data availability • Online open knowledge base of who is doing what and where
• Virtual and real-world knowledge events
• Donors to fund high-resolution EO data for use by humanitarian organisations
Ethics & privacy ocument and share ethics and privacy best practices
• Document
• Upscale community involvement in data generation through community mapping and validation
exercises
Technical expertise & skills • A playbook of principles and case studies that showcase good practice
• Su
Support regional centres of expertise
Financing for application ppropriate patient capital from development agencies and philanthropic organisations through the
• Appropriate
development & scaling creation of new financing mechanisms
cr
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Procurement challenges ining for demand-side organisations on satellite application procurement, processes, and pricing
• Training
• Mechanisms to aggregate demand into bulk procurement agreements across multiple demand-side
organisations
or
Piloting & duplication rmation of who is doing what and where via an online knowledge base
• Information
• Virtual and real-world knowledge events
• Enc
Encouraging shared access to data
Conclusions
8
Background
Background
Key points
• The UK Humanitarian Innovation Hub is a pilot programme, funded by the
UK Foreign, Commonwealth and Development Office (FCDO) and hosted by
Elrha, that seeks to identify the biggest challenges within and opportunities to
address global humanitarian responses.
• The objective of this report is to provide a consolidated view on the current use
of satellite applications in humanitarian settings.
• Satellite applications are digital services and products that serve several
functions for society, the environment, and the economy by deploying
three types of satellite technology: Satellite Earth Observation, Satellite
Communications, and Global Navigation Satellite Systems.
Satellite applications
Satellite applications are digital services and products that serve several functions
for society, the environment, and the economy by deploying three types of satellite
technology.
Earth Observation (EO) is the gathering of information about the physical, chemical,
and biological systems of the planet via remote-sensing technologies. EO is used to
monitor and assess the status of, and changes within, the earth’s atmosphere, oceans,
land surface, and built environments.
Satellite Communications
11
Humanitarian
emergencies
Humanitarian emergencies
Key points
• Humanitarian emergencies are “events or series of events that represent a
critical threat to the health, safety, security, or wellbeing of a community or
other large group of people.” They are widely geographically distributed, often
covering large areas and multiple populations. They are diverse and complex,
and require coordinated efforts and resources across a variety of local, national,
and international stakeholders.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
They are broad and diverse; whilst no two humanitarian emergencies are the same, they
can share common characteristics. They are based around an event in time that can be
either slow to emerge (e.g., drought) or rapid (e.g., a flood). They pose a threat to large
groups of people. There are typically well-defined phases of an emergency or disaster
management situation that follow a life-cycle process.
13
• Determine the high-level scope for this report in terms of which existing “domains,”
e.g., food insecurity, and disasters, should be included in scope for humanitarian
emergencies.4
• Identify key stakeholders and their needs in the phases of a humanitarian emergency.
Domains are used to illustrate and define the scope for the report. Domains are simpler
but less specific than the sector and cluster groupings often used by humanitarian
organisations, such as the sphere response sectors and UN Inter-Agency Standing
Committee (IASC) clusters.5 Domains are often closely related and more than one
may apply to an individual humanitarian emergency. This section describes the most
common domains in humanitarian response where satellite technology has the potential
to support. For the purposes of the research, the domains are shown in Figure 1.
14
Security and conflict: When the protection of the territorial integrity, stability, and
vital interests of states through the use of political, legal, or coercive instruments
at the state or international level fails, violent conflict and insecurity result. Violent
conflicts lead to death and destruction, the crumbling of weak states, local and
international insecurity, a vicious cycle of underdevelopment, instability, and large-scale
humanitarian crises.7
Food insecurity: Food security means that all people, at all times, have physical, social,
and economic access to sufficient, safe, and nutritious food that meets their preferences
and dietary needs for an active and healthy life. Natural hazards, conflict, and supply
chain inefficiencies can disrupt access and contribute to food insecurity. Existing food
insecurity is often exacerbated by conflict that displaces communities that are unable to
carry out day-to-day activities, such as managing crop fields.
million people are refugees, 68% of whom are from Syria, Venezuela, Afghanistan,
South Sudan, and Myanmar. Further, 39% of refugees are hosted in five countries:
Turkey, Colombia, Pakistan, Uganda, and Germany.10 Internally displaced persons
(IDPs) are those forced to flee their homes, but stay within their own country. IDPs
also include people displaced by internal strife and natural disasters. Over 46 million
people are estimated to be IDPs.11
6 For example, 43% of all disasters between 1995 and 2015 in the CRED dataset were flooding-related; Asia and Africa were most at risk. Beatriz
Ravilla Romero, ‘Improving Global Flood Forecasting using Satellite Detected Flood Extent’ (PhD diss., Utrecht University, 2016); Centre for
Research on the Epidemiology of Disasters, The Human Cost of Weather Related Disasters, 1995–2015, 2015, www.unisdr.org/2015/docs/climatechange/
COP21_WeatherDisastersReport_2015_FINAL.pdf
15
7 UN International Peace Academy, The Security-Development Nexus: Conflict, Peace and Development in the 21st Century, May 2004,
www.ipinst.org/wp-content/uploads/publications/security_dev_nexus.pdf
8 UN High Commissioner for Refugees, ‘Refugee Data Finder,’ updated 18 June 2021, www.unhcr.org/refugee-statistics/
9 Same as above.
10 Same as above.
11 UN High Commissioner for Refugees, ‘Refugee Facts: What Is a Refugee?,’ www.unrefugees.org/refugee-facts/what-is-a-refugee/
12 US Centers for Disease Control and Prevention, ‘COVID-19: Identifying the Source of the Outbreak,’ updated 1 July 2020,
www.cdc.gov/coronavirus/2019-ncov/science/about-epidemiology/identifying-source-outbreak.html
Humanitarian events
A humanitarian event is a potential starting point for a humanitarian emergency and
the need for humanitarian action. It is a specific occurrence that initiates the need for
decision-making or action by stakeholders within identified common processes. This
report is concerned with humanitarian events that encompass one or more hazards
that may precipitate a humanitarian crisis and have significant consequences for those
affected. The term “event” is used broadly to describe circumstances wider than hazards,
which can have context-specific meanings.
The research framework includes events and hazards in the identified domains rather
than an exhaustive review. Humanitarian events are rarely discrete and are often
the result of other related events; for example, an IDP crisis may be precipitated by
conflict or a disaster. The exact definition of scope for a specific event and, ultimately,
responsibility for the response to an event is the role of one or more stakeholders. This
may vary by geography and organisational structure, but this report attempts to identify
commonalities across events. It is increasingly common practice to name individual
instances of a realised hazard to aid communications, e.g., the naming of individual
cyclones by meteorological services. Due to the interrelated and combined nature of
events, the term event cluster is used below.
Hazards
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Humanitarian events include newly emerged and changing hazards. Hazards are
often described as natural (e.g., fire) or technological (e.g., radiation), with a socio-
natural hazard (e.g., increased flood risk due to environmental degradation) being a
combination of the two. Caribou Space also noted the use of “Natech,” which refers to a
technological hazard caused by a natural one.15 Hazards are global in nature, occurring
with varying severity and frequency in all countries.
Time is an important factor for events, hazards, and their resulting crises and disasters.
They may be permanent, temporary, or episodic.16
• Permanent hazards, for the purposes of risk management, are always present
(e.g., coastal erosion and climate change).
• Temporary hazards can be long-duration (e.g., slow onset) but have a clear start and end
Humanitarian emergencies
• Episodic hazards tend to be shorter in duration but recurring (e.g., hurricanes, winter
storms, tsunamis, and river floods).
16
Hazards or events that have not exceeded defined risk thresholds may require
monitoring to determine if they will. Some events may be impossible, or at least
difficult, to identify as hazardous in advance with sufficient detail to mitigate, e.g.,
long-term consequences of a political decision.
Existing typologies and records of hazards and disasters, such as the Centre for
Research on the Epidemiology of Disasters (CRED) Emergency Events Database
(EM-DAT), are essential datasets and schemas.17 However, they are typically more
detailed and granular than those found in common usage by humanitarian actors
and stakeholders. The event types used in this report are derived from the existing
typologies within the UNDRR Hazard Clusters.18
These events relate well to the “disasters” domain, but less so to the other four domains.
Therefore, other sources such as The Sphere Handbook have also been used.19 Caribou
Space do not seek to redefine hazard or event types, but rather to highlight those
practically applicable to this research; therefore, some omission and consolidation have
been made. Omissions include cyber hazards, mental health, and food safety, as these
were felt to be outside the direct scope of satellite tools to address.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Risk
For this report risk means “the probability of an outcome having a negative effect on people,
systems or assets.” 20 We also note the importance of valuing potential outcomes from
risks where measured value may be less tangible (e.g., sustainability) or measured and
defined differently by separate stakeholders.21
importance of also addressing multiple hazards and interrelated and dynamic changes
to risk is increasingly recognised.25
17 Centre for Research on the Epidemiology of Disasters, ‘EM-DAT: The International Disaster Database,’ https://emdat.be
18 UNDRR International Science Council, Hazard Definition and Classification Review, 2020, https://www.undrr.org/publication/hazard-definition-
and-classification-review
19 The Sphere Project, The Sphere Handbook 2018, 6 November 2018, https://spherestandards.org/handbook-2018/
17
20 UNDRR, ‘Understanding risk,’ https://www.undrr.org/building-risk-knowledge/understanding-risk
21 Baruch Fischhoff and John David Kadvany, Risk: A Very Short Introduction (Oxford University Press, 2011)
22 Catriona Mackenzie, Wendy Rogers, and Susan Dodds, eds., Vulnerability: New Essays in Ethics and Feminist Philosophy (Oxford University Press, 2013)
23 Jake Ansell and Frank Wharton, eds., Risk: Analysis, Assessment and Management (Wiley, 1992)
24 UNDRR, Global Assessment Report on Disaster Risk Reduction 2019, 2019, www.undrr.org/publication/global-assessment-report-disaster-risk-
reduction-2019
25 Melanie S. Kappes, Margreth Keiler, Kirsten von Elverfeldt, and Thomas Glade, ‘Challenges of Analyzing Multi-Hazard Risk: A Review,’ Natural
Hazards 64 (2012): 1925–58, https://doi.org/10.1007/s11069-012-0294-2
Compound risks and complex emergencies
This section has focused on introducing a research framework to outline the wider
humanitarian context and define the scope of the research in terms of priority domains
and hazards.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Humanitarian emergencies
18
26 A. Kruczkiewicz, et al., ‘Opinion: Compound risks and complex emergencies require new approaches to preparedness,’ PNAS 118, no. 19 (2021):
article e2106795118, https://doi.org/10.1073/pnas.2106795118
27 UNHCR, ‘Coordination in Complex Emergencies,’ 1 September 2001, https://www.unhcr.org/en-us/partners/partners/3ba88e7c6/coordination-
complex-emergencies.html
Environmental and
situational data for
humanitarian
decision-making
Environmental and situational data
for humanitarian decision-making
Environmental and situational data for humanitarian decision-making Beyond Borders: Satellite Applications for Humanitarian Emergencies
Key points
• Satellite applications, as digital technology, provide a unique source of data
about people and their surrounding environment—with the potential to
fill significant data and information gaps in humanitarian management and
decision-making processes.
• The World Bank’s Data for Better Lives report highlights how the production
and collection of data enable humanitarian and development outcomes.
The World Bank’s Data for Better Lives highlights how the production and collection
of data enables humanitarian and development outcomes. Individuals, civil society, and
academia gain greater accountability and government and international organisations
improve policymaking and service delivery, whilst the private sector gain increased
business opportunities. However, these positive outcomes come with the risks of greater
criminal activity, political surveillance, market concentration, widening inequality, and
discrimination.
20
DATA ANALYTICS
AND PROCESSING
Reuse
Reuse
Figure 2: World Bank: How Data Can Support Development: A Theory of Change28
Humanitarian actors continually make complex decisions in the preparation for, and
response to, humanitarian emergencies. Digital technologies support these decision-
making processes by providing large volumes of high-quality data that enables insight.
The term “digital data for development (D4D)” refers to the use of data generated by
mobile, satellite, and digital devices to inform and strengthen nonprofit, humanitarian,
and public-sector decision-making.29
21
28 World Bank, World Development Report 2021 Data for Better Lives, 2021, www.worldbank.org/en/publication/wdr2021
29 igital Impact Alliance, Leveraging Data for Development to Achieve Your Triple Bottom Line, July 2018, http://digitalimpactalliance.org/wp-content/
D
uploads/2018/07/DIAL_D4D_MNO-Paper2_2018-v9.pdf
30 London Economics, Economic evaluation of the International Partnership Programme (IPP): Cost-Effectiveness Analysis, October 2019,
https://londoneconomics.co.uk/blog/publication/economic-evaluation-of-the-international-partnership-programme-ipp-cost-effectiveness-
analysis-october-2019/
Data from satellite applications has the specific advantages of:
• Coverage: Satellites have global coverage that makes it possible to consistently monitor
vast, remote, and even conflict regions across countries and continents.
• Repeatability: Satellite observations are collected along a periodic orbit of the Earth’s
surface, so they are repeatable and comparable over time.
• Continuity: The continuity of satellite data streams allows time to build experience and
refine the systems that use the data.
Environmental and situational data for humanitarian decision-making Beyond Borders: Satellite Applications for Humanitarian Emergencies
• Thematic detail: The range of satellite sensors now available allows for application to a
wide range of humanitarian domains.
• Speed: Increasingly, satellite data is available for use soon (days or even hours) after it
is acquired, enabling stakeholders to receive the derived information they need to act
quickly—critical in, for example, disaster scenarios.
• Affordability: Along with the increase in commercial satellites, there is also an increase
in satellites that allow free and open access to data, such as the Copernicus Sentinel
missions.
22
Ecosystem of satellite
applications for
humanitarian
emergencies
Ecosystem of satellite applications
for humanitarian emergencies
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Key points
• There is a complex network of stakeholders involved in satellite applications
for humanitarian emergencies; the report provides a comprehensive list of
actors on this topic.
− Academia
Ecosystem of satellite applications for humanitarian emergencies
− NGOs
− Media
− Development Agencies
− Governments
− First Responders
− Private Sector
− Academia
− NGOs
− Media
− Development Agencies
− Affected Public
24
“SUPPLY-SIDE” ACTORS “DEMAND-SIDE” ACTORS
PRIVATE SUPPLIER
Other
Development Agency Media General Public
PUBLIC SUPPLIER
Private Sector
Satellite Operator
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Analytics Provider
All stakeholders related to the use of satellite applications for humanitarian emergencies
can be positioned along a spectrum from the supply-side to the demand-side. The
supply-side provides the satellite applications, whereas the demand-side uses those
applications to address their humanitarian emergencies.
Supply-side actors
The humanitarian ecosystem is composed of an inherently heterogeneous mix of
traditional actors (NGOs and Governments) as well as non-traditional actors (voluntary
Ecosystem of satellite applications for humanitarian emergencies
Each stakeholder group has a particular set of interests, motives, assets, skills, and
capacities which directly influence their use of data and value derived from these assets.
Humanitarian actors, their partners, and the Affected Public are producing, capturing,
and accessing growing volumes of data about vulnerable populations and their
environments. These data providers and users together make up a diverse and complex
“humanitarian data ecosystem.”
25
Supply-side actors
Private Suppliers and Public Suppliers represent the majority of providers of satellite
applications for humanitarian emergencies.
Private Suppliers are a diverse group that is further subdivided. Satellite Operators and
Resellers, such as Inmarsat (SatComms) and Maxar (EO), own and operate satellites
and sell data streams from them, either directly or via resellers. Cloud Computing
Providers, such as Google Cloud Platform, Amazon Web Services, and Microsoft
Azure, make it possible to process this data into products by providing on-demand
cloud infrastructure. This software as a service and infrastructure enables all data
analysis and product generation to be implemented in the cloud instead of the user’s
desktop.31 Google Earth Engine (GEE), a cloud-based platform, combines a multi-
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Public Suppliers include two subgroups: first, Satellite Operators, which includes
government space agencies such as the European Space Agency (on behalf of the
European Commission) with their Copernicus constellation and the US National
Aeronautics and Space Administration (NASA). Second, Analytics Providers, such as
mapping agencies, statistical bureaus, land ministries, health ministries, defence and
intelligence agencies, and disaster response agencies use satellite data alongside other
data sources to create and contribute analyses and reports.
Academia and NGOs play roles on the supply-side as data providers and on the
Ecosystem of satellite applications for humanitarian emergencies
Media often source and use satellite imagery in news articles to raise awareness among
the public about humanitarian emergencies.
Development Agencies are occasionally on the supply-side, such as the World Bank or
other International Financial Institutions (IFIs), and have internal specialist teams able
to produce satellite applications for humanitarian contexts.
26
31 Catherine Nakalembe et al., ‘A review of satellite-based global agricultural monitoring systems available for Africa,’ Global Food Security 29
( June 2021): article 100543, https://doi.org/10.1016/j.gfs.2021.100543
32 Noel Gorelick et al., ‘Google Earth Engine: Planetary-scale geospatial analysis for everyone,’ Remote Sensing of Environment 202 (December 2017): 18–27,
https://doi.org/10.1016/j.rse.2017.06.031
table 1: Supply-Side Stakeholder Groups
Satellite • Airbus, Capella Space, Earth-i, GHGSat Inc., Planet Labs, Inmarsat, Maxar Technologies,
Operators OneWeb, Satellogic, Telespazio
& Resellers
Cloud • Alibaba Cloud, Amazon Web Services, Google Cloud Platform, IBM Cloud, Microsoft
Computing Azure, Oracle, Salesforce, SAP Cloud Platform
Providers
Platform/ • Google Earth Engine, Microsoft Planetary Computer, Orbital Insight, Descartes Labs,
Private Suppliers Solution CrowdAI, Ecometrica, MapBox
Providers
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Hardware/ • Hardware: Garmin, Thales Alenia Space, Gilat Satellite Networks, Airbus
Software • Software: Harris (ENVI), ESRI, PCI Geomatica
Suppliers
• Open access: QGIS
• Research institutes: Norwegian Centre for Humanitarian Studies, Canadian Research Institute on
Humanitarian Crisis and Aid (OCCAH), African Centres for Disease Control
Academia
• Universities: University of Geneva, Harvard Humanitarian Initiative, Nanyang Technological University
• Think tanks: Chatham House, Freedom House, Center for Strategic International Studies (CSIS)
Ecosystem of satellite applications for humanitarian emergencies
Development Agency • International Financial Institutions: World Bank, Asian Development Bank
27
Demand-side actors
The diverse community of “data providers” is met by an even more diverse community
of global humanitarian data users. Many of these humanitarian actors focus on specific
domains or event types, as discussed in the previous chapter, while others are better
categorised by their roles and workflows during an event.
Ordnance Survey or Survey of India, and defence and intelligence, e.g., Ministries of
Defence. They provide other parts of government with analysis to support decision-
making. They also produce detailed reports for media briefings.
First Responders include emergency services, military response units, and in-country
aid agencies. They provide support in terms of evacuations, search and rescue, and
emergency food, shelter, and medical supplies.
Media include local and international journalists. They report on the humanitarian
emergency to the broader public.
Private Sector includes insurance and reinsurance firms that provide cover against asset
losses, e.g., infrastructure, vehicles, or crops, in the event of a humanitarian emergency,
e.g., a disaster. It also includes multinational companies like Coca-Cola and Nestlé that
have supply chains across the world, which are affected by humanitarian events.
Academia plays roles on both the supply-side as data providers and also as demand-side
Ecosystem of satellite applications for humanitarian emergencies
users.
NGOs include local and international NGOs and human rights organisations. They
often provide on-the-ground support in the case of humanitarian events, including
providing basic needs such as shelter, water, and healthcare.
Affected Public includes affected communities who play a role in working with
humanitarian stakeholders, especially those on the front lines, to direct and prioritise
assistance.
General Public includes the local and international public. They are government
constituents and influencers, and in some cases contribute to philanthropic funding
appeals to support NGOs in their humanitarian response.
28
table 2: Demand-Side Stakeholder Groups
Humanitarian
Stakeholder Group • Examples
Government • Ministers
• National coordinators: National Disaster Risk Reduction & Management Council (NDRRMC),
Civil Contingencies Secretariat, Pacific Community (SPC)
• Regional: African Risk Capacity
• Mapping agencies: Ordnance Survey, Survey of India, National Geographic Department of Lao (NGD)
• Defence and intelligence: Ministries of Defence, United States Africa Command (AFRICOM)
Development Agencies • Bilaterals and multilaterals: UK Foreign, Commonwealth and Development Office (FCDO),
Japan International Cooperation Agency (JICA)
• IFIs: World Bank, International Finance Corporation (IFC), Asian Development Bank,
International Development Bank (IDB)
• Private foundations: Bill and Melinda Gates Foundation, Rockefeller Foundation
• Corporate philanthropies: Mastercard Foundation
Affected Public • Communities directly affected by humanitarian events, who are also volunteers and
crowdsourcing contributors
• Vulnerable groups like refugees and IDPs, slum dwellers, farmers, children, etc.
General Public • Broader national or international public following news of the humanitarian event
• Volunteers and crowdsourcing contributors, e.g., OpenStreetMap volunteers
Ecosystem of satellite applications for humanitarian emergencies
29
Humanitarian data roles
Beyond thinking about these actors by their traditional stakeholder groupings, e.g.,
Government, Development Agency, etc., it can also be helpful to understand these
groups vis-à-vis their interactions with data. In their paper, Haak et al. present four
different stakeholder data-oriented roles involved in developing, disseminating, and
using a “toolbox” of information derived from public and private sources of data:33
• Toolbox Users are humanitarian or government agencies who should use the toolbox
(i.e., the data processed into understandable information) and incorporate the outputs
into their operations and decision-making.
In some cases, a given stakeholder group can play multiple roles. For example,
governments are often Local Leads, Data Suppliers, and Toolbox Users; their primary
data roles usually vary across departments or agencies. As Local Leads, governments
are on the front lines of a humanitarian crisis and coordinating response efforts. As
Data Suppliers, governments have made significant investments in providing free and
open, continuous and consistent satellite data streams.
In recent years, these governments and the agencies that design and operate satellites
have more actively engaged with the broader humanitarian community to ensure
Ecosystem of satellite applications for humanitarian emergencies
As Toolbox Users, government agencies leverage public and private satellite data
alongside other sources to inform their analyses, findings, and reports. For example,
national statistical offices use satellite imagery to streamline enumeration campaigns
and more efficiently complete censuses.35
30
33 Elise Haak et al., ‘A framework for strengthening data ecosystems to serve humanitarian purposes,’ Proceedings of the 19th Annual International
Conference on Digital Government Research: Governance in the Data Age (May 2018): article 85, https://doi.org/10.1145/3209281.3209326
34 US Geological Survey, ‘Landsat’s Economic Value to the Nation Continues to Increase,’ 4 March 2021, www.usgs.gov/center-news/landsat-s-
economic-value-nation-continues-increase; European Commission, Copernicus Market Report, February 2019, www.copernicus.eu/sites/default/files/
PwC_Copernicus_Market_Report_2019.pdf
35 Sarchil Qader, ‘Semi-automatic mapping of pre-census enumeration areas and population sampling frames,’ Humanities and Social Sciences
Communications 8 (2021): article 3, https://doi.org/10.1057/s41599-020-00670-0
Enabling environment
Surrounding the supply- and demand-side is an environment that is required to enable
and make best use of satellite applications. Components of the enabling environment
are listed below and detailed later (see section Overcoming barriers to uptake of satellite
applications in humanitarian emergencies):
• Education and skills encompasses the general awareness, domain expertise, and
technical skills needed to produce or use satellite applications for humanitarian
Beyond Borders: Satellite Applications for Humanitarian Emergencies
applications.
• Finance involves the financial organisations providing funding to both supply- and
demand-side organisations, and to other parts of the enabling environment, to facilitate
the development and use of satellite applications.
• Digital/IT infrastructure accounts for the associated technologies and data required to
leverage satellite data streams, including but not limited to cloud computing, internet
connectivity, ground stations, geospatial software, mobile applications and tools, and
other data sources to enhance satellite-derived insights.
Dig
tion ita
Ecosystem of satellite applications for humanitarian emergencies
l In
gula fr
e
r
as
&
tru
icy
ctu
Pol
re
e
nc
ti o
n
& na
sk
ill s Fi
31
Satellite
applications
Satellite applications
Key points
• The most persistent and extensive humanitarian data gaps include a lack of
information on affected schools, malnutrition, damaged infrastructure like
buildings and roads, and refugees, internally displaced persons, and persons
of concern.
• Satellites are valuable in addressing data gaps across all of these areas for
every humanitarian emergency.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
• Satellite applications support all stages of the event life cycle, from Hazard
and Risk Assessment to Recovery. Satellite EO can be particularly powerful
in supporting predictive analytics for slow-developing crises, for example,
in the context of food insecurity predicting weak crop yields months before
the harvest season, or to anticipate violence and understand the drivers of
conflict.
health and education were the least complete categories. Figure 5 plots HDX’s data
gaps against the most relevant satellite-enabled use case from those listed in this report.
33
36 Sarah Telford, ‘The State of Open Humanitarian Data,’ OCHA Centre for Humanitarian Data, 10 February 2021, https://centre.humdata.org/the-
state-of-open-humanitarian-data-2021/
Figure 5: Completeness of Data Required Across Humanitarian Operations (left) Mapped to Satellite Use Case (right)
11 63 26
Humanitarian Needs Population and poverty, & Hazardous event detection and severity
74 7 19
Health Facilities Infrastructure mapping and exposure, & Microplanning for health interventions
22 78
Roads Infrastructure mapping and exposure, & Microplanning for health interventions
Satellite applications
41 59
Conflict Events Hazardous event detection and severity, & Evidencing conflict-related violence
96 4
Data on severe acute malnutrition rates is also largely unavailable. However, combining
detection, extent, and severity of events with population and poverty mapping and
forecasts of crop yield can improve these estimates. While satellites cannot directly
observe nutrition levels, they can offer useful proxy information that, when combined
with other data sources, closes some of this data gap. While there has been extensive
usage of satellite imagery for crop models, a focus on nutrition is relatively early-stage
research.37
Across most data categories, there were uniform calls for improving data quality, as well
as access to persistent, real-time data and information streams.38 Beyond types of data
needed, the following requirements emerged from this study, many of which are closely
related:
• Predictive analytics42
Needs assessments are critical during humanitarian crises, as they provide baseline
information (e.g., the number of people affected) upon which situation monitoring,
response planning, and severity assessments are based. Needs assessments provide
evidence for the strategic decisions that need to be made, such as where to allocate or
where to deliver aid. In some cases, the availability of such data has been constrained
due to limited access to insecure and remote areas. Moreover, many governments do not
have access to adequate data on their entire populations. This is particularly true for the
poorest and most marginalised communities. Satellites can help provide comprehensive
population data regardless of insecurity or distance from urban centres.
Satellite applications
37 Jeremy Irvin, Dillon Laird, and Pranav Rajpurkar, ‘Using Satellite Imagery to Predict Health,’ 2017, http://cs231n.stanford.edu/reports/2017/pdfs/559.pdf;
Elizabeth Bondi et al., ‘Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency,’ KDD 2020 Workshop on
Humanitarian Mapping, 2020, https://teamcore.seas.harvard.edu/files/teamcore/files/kdd_humanitarian_mapping_workshop_2020_4.pdf
38 Henry Lewis and Gary Forster, Humanitarian Research Brief 4: Data Use Capacity in Protracted Humanitarian Crises, Humanitarian Data
Transparency Series Brief 4, Publish What You Fund, June 2020, https://www.publishwhatyoufund.org/wp-content/uploads/dlm_uploads/2020/06/
Humanitarian-Research-Brief-4.pdf
39 Same as above; United Nations, ‘Big Data for Sustainable Development,’ www.un.org/en/global-issues/big-data-for-sustainable-development
40 Kees Boersma and Chiara Fonio, eds., Big Data, Surveillance and Crisis Management (Routledge, 2017); Tina Comes, Kenny Meesters, and Stina
35 Torjesen, ‘Making sense of crises: The implications of information asymmetries for resilience and social justice in disaster-ridden communities,’
Sustainable and Resilient Infrastructure 4, no. 3 (2019): 124–36, https://www.researchgate.net/publication/321691714_Making_sense_of_crises_the_
implications_of_information_asymmetries_for_resilience_and_social_justice_in_disaster-ridden_communities
41 Comes, Meesters, and Torjesen, ‘Making sense of crises.’
42 UN Committee of Experts on Global Geospatial Information Management, Future Trends in Geospatial Information Management: The Five to Ten
Year Vision, August 2020, https://ggim.un.org/documents/DRAFT_Future_Trends_report_3rd_edition.pdf; Kevin Hernandez and Tony Roberts,
Predictive Analytics in Humanitarian Action: A Preliminary Mapping and Analysis, K4D Emerging Issues Report 33, Institute of Development
Studies, June 2020, https://opendocs.ids.ac.uk/opendocs/bitstream/handle/20.500.12413/15455/EIR33_Humanitarian_Predictive_Analytics.
pdf ?sequence=1&isAllowed=y
Democratisation of access to data, especially for affected communities, ensures
that humanitarians can better collect data, share information, and validate findings
with communities on the ground. This can inform intervention priorities and
locations to target resources and services. In a context of increasing social and digital
inequalities, technologies run the risk of continuing to benefit more resilient and less
at-risk populations. However, with intentional humanitarian data stewards—people
responsible for championing improved data usage—satellites can help address digital
divides during events rather than exacerbate them. For example, satellite imagery
provides context and maps of every community regardless of geographic spread, size,
wealth, political affiliations, or administrative boundaries. By enabling feedback loops
with affected communities, humanitarians would not only ensure the right to access the
data collected about these populations but also gain the best possible “groundtruth,” or
confirmation of their interpretation of remotely collected data.
and evaluation.
36
Use of satellite applications across the event life cycle
Satellite applications have a role across all stages of a humanitarian emergency, which
are visualised in Figure 6.
1 Hazard and Risk Assessment – The first stage whereby hazards and risk factors
that have the potential to lead to a humanitarian situation are identified. Hazard
identification involves characterising hazards according to the probability of the hazard
becoming reality and its potential severity.
2 Mitigation – This stage includes any action or sustained effort undertaken to reduce a
hazard through the reduction of the likelihood and/or the consequence component of
that hazard’s risk.
4 Response – The following stage involves actions aimed at limiting injuries, loss of life,
and damage to property and the environment taken before, during, and immediately
after an event. Response processes begin as soon as it becomes apparent that an event is
imminent and last until the emergency is declared to be over.43
5 Recovery – The stage when communities, systems, and lives are rebuilt, repaired, and
otherwise returned to functional conditions.
Table 3 provides use case examples categorised by the humanitarian event life cycle.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
These examples are not exhaustive but are indicative of the wide-ranging and persistent
applicability of satellite data across the humanitarian spectrum.
Each use case is categorised as to whether satellite data provides direct value in
providing insights or indirect value because it needs to be combined with other data,
improved with ground validation, and/or integrated into a more sophisticated analysis
or model. For example, predicting and analysing crop yields remotely is a powerful use
case of EO in the humanitarian context, for technical reasons and given the potential
impact on saving lives. In turn, poverty mapping is of course important for the creation
of baselines, but EO can only effectively help in this case if combined with other
datasets and (often complex) analyses, and faces caveats and margins of error.
Haza
ery rd a
cov nd
Re Ri
s
kA
sse
ssm
Satellite applications
ent
onse
Resp
n
a ti o
i ti g
M
37
P re
p ar e d n e ss
43 Damon P. Coppola, Introduction to International Disaster Management, 3rd ed. (Elsevier Science, 2007)
table 3: List of Use Cases by Event Life Cycle Phase
Crop yield and • Early warning systems for crop failure or seasonal decline
productivity Direct value
forecasting • Yield gaps based on potential and observed crop growth
Election polling site • Preventing electoral violence with monitoring of election sites
Direct value
monitoring • Ensuring vulnerable populations have access to polling sites
Preparedness
Micro-planning for • Guiding immunisation campaigns and distribution of vaccines
Indirect value
health interventions • Navigating community health workers to reach every household
38
Agricultural pest and • Monitoring conditions on-farm and assessing risk for pests and disease
Direct value
disease response • Allocating limited supplies (e.g., pesticides, sprayers) to affected areas
Search and rescue • Identifying indications of human activity or presence along with
Direct value
missions terrain models to focus search areas
Access routes and • Identifying traversable routes based on debris and roadblocks
Direct value
evacuation plans • Assessing access to markets and trade hubs in rainy season
Response
Public crisis • Evidencing humanitarian need for fundraising appeals
Direct value
awareness • Using imagery in news stories to provide context
Humanitarian
stakeholder Direct value • Providing situational awareness to enable stakeholders to coordinate
coordination
Disease prevalence
Indirect value • Predicting the location and spread of diseases
and transmission
Analysing refugee
resettlement Indirect value • Identifying and planning locations to resettle refugees
suitability
39
Use cases by domain
Satellites provide critical data across all domains of humanitarian activity.
This section provides a broad overview of use cases by humanitarian domain.
Disasters
Emergency response
2.
5 Damage assessment
2.8 % 27.2%
% Subsidence and erosion
3.3
% Infrastructure mapping and exposure
Humanitarian stakeholder coordination
4.2%
Disaster risk financing and insurance
4.7% Natural resource mapping
Population and poverty
5.0% Contaminated sites
13.6% Infrastructure reconstruction — 1.9%
6.1% Crop/livestock insurance — 1.4%
Connectivity for public or — 1.4%
8.1% remote populations
12.2%
Extraction and trafficking of — 1.1%
illegal resources or humans
Other
Satellites are valuable assets in conflict situations, where access to the affected area
is often problematic for ground teams aiming to conduct conventional field mapping
40
during early phases. Without a physical presence on the ground, satellites can provide
timely intelligence on conflict-related destruction, displaced populations, refugee
and IDP camp growth, and human rights abuses, as well as inform post-conflict
reconstruction and monitoring of peace treaties.
Evidencing conflict-related violence
3.5%
Extraction and trafficking of illegal
15.4% resources or humans
2.
2.
1
1%
Damage assessment
2.8 %
% Monitoring prison camps
2.8
% Public crisis awareness
Infrastructure mapping and exposure
4.9% 14.0%
Infrastructure reconstruction
Nuclear facilities
6.3%
Natural resource mapping
Election polling site monitoring
6.3% Compliance with peace treaties
13.3%
Refugee camps
6.3%
Search and rescue missions
7.0% 7.7%
Population and poverty — 1.4%
Monitoring and evaluation — 1.4%
Other
In some cases, satellites can also help prevent conflict by evidencing de-escalation of
militarisation and adherence to international humanitarian law. As documentation
Beyond Borders: Satellite Applications for Humanitarian Emergencies
tools, satellites have the potential to collect important real-time evidence for alleged
war crimes and crimes against humanity. Because these events often occur in
inaccessible environments, over large geographic areas, and/or across long and multiple
time frames, high-resolution satellite imagery, in particular, has value for international
tribunals investigating these crimes.
Food insecurity
In the mitigation phase, satellite data, including both EO and SatComms, can inform
the implementation of disaster risk financing, agricultural insurance, and access to
credit, supporting more resilience in communities especially vulnerable to climate
shocks and stressors.
41
6.0% Crop yield and productivity forecasting
Natural resource mapping
Population displacement
Owing to its continuous spatial coverage and frequency of collection, satellite data, like
Beyond Borders: Satellite Applications for Humanitarian Emergencies
optical and synthetic aperture radar (SAR), provides insightful, near-real-time context
for analyses.44 By linking this data with GNSS data, population displacement can be
mapped and monitored.
Data derived from optical imagery, like temporary settlements, cattle, boats, vegetation,
water, and other features, can be used to estimate the number of people displaced or
the extent and status of infrastructure and natural resources in resettlement areas. It
also provides context-specific natural hazard risk assessment for informing mitigation
measures to protect already vulnerable communities.45
various biophysical and socioeconomic indices, can help build machine learning models to
improve forecasting of population displacement and simulate optimal response scenarios.
42
44 International Organization for Migration, ‘Migration Data Portal,’ updated 26 May 2021, https://migrationdataportal.org/data-innovation/
interactive-tool-map-world-internally-displaced-satellite-imagery
45 Nistara Randhawa et al., ‘Fine scale infectious disease modeling using satellite-derived data,’ Scientific Reports 11 (2021): article 6946,
https://doi.org/10.1038/s41598-021-86124-2; Shifeng Wang, Emily So, and Pete Smith, ‘Detecting tents to estimate the displaced populations of
post-disaster relief using high resolution satellite imagery,’ International Journal of Applied Earth Observation and Geoinformation 36 (April 2015):
87–93, https://doi.org/10.1016/j.jag.2014.11.013; A. Heslin, L. Thalheimer, and M. J. Purma, ‘Using Remote Sensing to Measure Displacement and
Assess Environmental Conditions at Host Locations’ abstract, American Geophysical Union Fall Meeting 2019, https://ui.adsabs.harvard.edu/
abs/2019AGUFMGC13G1225H/abstract
Refugee camps
Population and poverty
Public crisis awarenesss
Infrastructure mapping and exposure
Damage assessment
3.1 Evidencing conflict-related violence
% 35.9%
3.1% Emergency response — 1.6%
Natural resource mapping — 1.6%
4.7% Microplanning for health interventions — 1.6%
Extraction and trafficking of illegal — 1.6%
resources or humans
Connectivity for public or — 1.6%
14.1% remote populations
Contaminated sites — 1.6%
Monitoring and evaluation — 1.6%
23.4% Election polling site monitoring — 1.6%
Compliance with peace treaties — 1.6%
Refugee resettlement suitability analysis — 1.6%
Health emergencies
Beyond Borders: Satellite Applications for Humanitarian Emergencies
It is also possible to combine these data sources to forecast capacity and demand by
health clinics. Satellite applications can also be used to continuously assess national
health systems and generate health facility readiness scores for COVID-19 response.
While not specific to health, population and poverty mapping are critical to any health
surveillance approach, especially in an emergency setting where a lack of understanding
Satellite applications
of health catchment areas can cost precious time and resources. High-resolution
population estimates support the measurement of programme targets, coverage, and
improved programme planning.
Satellite data is valuable for mapping digital elevation models to chart navigation and
guide drones for vaccine delivery to remote clinics; however, this is a newer application,
43 as most cargo drones are still prototypes.46
46 Swiss Foundation for Mine Action, Drones in Humanitarian Action: A guide to the use of airborne systems in humanitarian crises, December 2016,
https://reliefweb.int/report/world/drones-humanitarian-action-guide-use-airborne-systems-humanitarian-crises
Microplanning for health interventions
2.7 %
Disease prevalence and transmission
2.7
2.
%
7% Population and poverty
As highlighted by population and poverty mapping, many use cases are valuable
across multiple domains and humanitarian events. For example, mapping damaged
infrastructure due to disaster or conflict can also inform better estimates of the
impacted and displaced populations and where food aid and other resources are
required. Other cross-cutting use cases include connectivity for public or remote
populations and early warning systems.
Satellite applications that support cross-cutting use cases are particularly relevant
for compound risks and complex emergencies. For example, in Lake Chad Basin
environmental degradation, a very high fertility rate, forced displacement, and repeated
herder–farmer conflicts create significant compound risk.47 EO is a powerful tool to
understand the dynamics and interrelation between such factors.
44
Key points
• Satellite Earth Observation (EO) is the gathering of information about the
physical, chemical, and biological systems of the planet via remote-sensing
technologies.
• Platform and cloud services, together with technologies like machine learning,
are simplifying access to and use of satellite applications.
Satellite applications can be a basic product or service from satellites, e.g., visual
imagery in a map to show urban infrastructure or satellite phone calls. Other
applications are the result of significant data processing and additional modelling,
e.g., asset exposure to floods within urban environments. For some applications, the
satellite data element may be small in comparison to overall activity. As introduced
earlier, suppliers are either Private Suppliers, Public Suppliers, Academia, NGOs,
Development Agencies, or a combination of these. Figure 12 describes the types of
46
services and products provided by suppliers.
Imagery Image & data Data integration
& data processing & interpretation
Satellite & ground Position, navigation Entity & object Location, intelligence Satellite
operations & timing signals location & analysis Applications
Global Navigation
Satellite System
Inmarsat Smart modules
& devices (IoT)
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
SatComms
Hiber
Figure 12: Components of Satellite Applications
For EO-based applications, basic imagery and its supply are core components. “Basic
imagery” is satellite imagery that has been acquired, downloaded, and made available
with minimal processing; this includes “raw” imagery straight from the satellite and
imagery that has been processed in several common ways, such as refining to give
a more precise location, transforming for measurement accuracy, and correcting for
various data artefacts that depend on the type of instrument or camera. These are
typically referred to as level one and level two imagery products.
47
“Analysis Ready Data” (ARD), an important development for comparative analysis
between imagery from different satellites, involves a higher level of processing. Whilst
not required for all applications, efforts like ARD decrease barriers to using multiple
sources of data and simplify the more rigorous analysis and scientific approaches
required for some applications.
An important attribute of imagery is its “resolution”: the smallest thing that can
typically be seen (resolved) on the ground. More correctly termed Ground Sample
Distance (GSD), this defines how much detail is available in imagery. Very High
Resolution (VHR) EO data is needed for many humanitarian use cases, particularly
when the emphasis shifts from broad environmental monitoring (e.g., forest change)
to observing features related to individuals and buildings (e.g., mapping informal
settlements). The resolution grades are listed below:
VHR data is more expensive than lower resolution data, as it is sourced from Private
Suppliers at commercial costs. Moreover, VHR data is also computationally more
expensive. Lower resolution imagery, such as 10 metres to 200 metres, is available from
Public Suppliers at low or no cost.
Coverage is a broad term that describes the amount of imagery and data collected
over an area or region with a given frequency. In general, Public Suppliers, such as the
European Commission’s Copernicus, have revisit times of five to eight days with global
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
coverage. Private Suppliers offer higher revisit times for smaller areas as standard.
For example, Planet provides new imagery at 0.5m resolution up to twelve times daily
from its SkySat constellation.48 Timely availability of data following a humanitarian
emergency (particularly in the Response Phase) is critical. Conversely, it is also
important to have access to fresh data for the Preparedness Phase, for example, in order
to quantify hazard risk scenarios. Imagery is often made available to humanitarian
organisations after an event but not preceding it.
The core delivery modes for satellite imagery are direct reception, tasked and repeat
imagery, archive imagery, and platform and cloud services, as detailed below.
Tasked and repeat imagery – Tasked imagery is the requesting of single or multiple
satellite images in the future. Procurers of these services typically specify a geographical
area (Area of Interest (AOI)), timeframe, and frequency of observations for imagery.
Cloud-free or low-cloud imagery may only be possible by preparing a mosaic of a
number of images acquired over a period of time. Imagery is often supplied via online
systems from satellite operators, distributors, and resellers through web and API access.
Satellites have capacity constraints; therefore, imagery needed for a given time and
location may not always be available. For some applications and imagery types, a prior
48
arrangement to task imagery is necessary to ensure that required imagery is collected by
satellites and made available.
48 Martin Van Ryswyck, ‘Planet Announces 50 Cm SkySat Imagery, Tasking Dashboard and Up To 12× Revisit,’ Planet, June 9, 2020,
www.planet.com/pulse/tasking-dashboard-50cm-12x-revisit-announcement/
Whilst tasking is still common, imagery from some providers is not tasked and provides
regular and routine “repeat” imaging of global regions according to a defined schedule.
This approach is common for large public sector satellites (e.g., USGS Landsat series
and EC Copernicus Sentinels) and increasingly for large constellations (e.g., Planet).
Archive imagery – Many Private Satellite Operators and Public Satellite Operators
make their recent and historical imagery available via online or offline archives. Some
suppliers opt to make imagery available as soon as it is collected, downloaded, and
processed, thus reducing or removing the need for tasked imagery. Public archives of
global satellite imagery are available, including ISRO49 (India), USGS50 and NASA 51
(USA), EC52 (Europe), JAXA 53 (Japan), and INPE54 (Brazil). Regular global imagery
at resolutions generally between 10m and 200m can be acquired from these archives.
The archives typically require registration and, whilst data access is generally free and
open, terms and conditions can apply. APIs for direct access via software and apps are
also commonly provided.
Private Satellite Operators provide online and API access to their archives for imagery
that generally has much higher resolution than that of public archives. Examples
include Maxar’s SecureWatch,55 Airbus’ OneAtlas,56 and Planet’s Platform.57 In many
cases tools for viewing and using the imagery are provided; they can be easy to use
but potentially limited in comparison to the more comprehensive cloud services or
platforms described below.
For some archives, users can experience delays between the collection of imagery
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
and its release due to data policy or commercial reasons that differentiate the service
from tasked imagery. In many cases, priority access is available in advance or by prior
arrangement for humanitarian events. Some commercial satellite operators offer free
data access for disasters and emergencies (e.g., Planet’s Disaster Data 58 and Maxar’s
Open Data 59 Program), and the Disaster Charter60 provides rapid access to nationally
mandated disaster management authorities upon request.
International actors within the EO community (CEOS and GEO) are encouraging the
use of open data cubes to host easily accessible datasets in a standard format ready for
use. Several projects are at various stages within several countries.61
Satellite imagery commonly requires processing and analysis in order to derive useful
information from it. For small numbers of images, this can often be performed on
workstations with freely available software. However, satellite imagery data tends to
require significant storage and computing power to routinely process and analyse large
datasets.
49 Indian Space Organisation National Remote Sensing Centre, ‘Bhuvan: Indian Geo-Platform of ISRO,’ https://bhuvan-app1.nrsc.gov.in/bhuvan2d/
50 US Geological Survey, ‘EarthExplorer,’ https://earthexplorer.usgs.gov
51 NASA Earthdata, ‘Earthdata Search,’ https://search.earthdata.nasa.gov/
52 ESA, ‘Copernicus Open Access Hub,’ https://scihub.copernicus.eu/dhus/#/home
53 Japan Aerospace Exploration Agency Earth Observation Center, ‘ALOS Global Digital Service Model,’
www.eorc.jaxa.jp/ALOS/aw3d30/l_map_v2003.htm
54 Instituto Nacional de Pesquisas Espaciais Earth Observation General Coordination, ‘Image Generation Division Catalogue,’
49 www.dgi.inpe.br/catalogo/
55 Maxar, ‘SecureWatch,’ www.maxar.com/products/securewatch
56 Airbus, ‘OneAtlas Data,’ www.intelligence-airbusds.com/imagery/oneatlas/data/
57 Planet, ‘Planet Platform,’ www.planet.com/products/platform/
58 Planet, ‘Disaster Data,’ www.planet.com/disasterdata/
59 Maxar, ‘Open Data Program,’ www.maxar.com/open-data
60 The International Charter Space and Major Disasters, https://disasterscharter.org/
61 Committee on Earth Observation Satellites, ‘CEOS Open Data Cube,’ www.opendatacube.org/ceos
Satellite imagery can be processed to derive physical attributes of geography within an
observed area, such as elevation,62 surface,63 and soil moisture.64 Dynamic natural and
man-made phenomena can be routinely mapped, such as fires,65 floods,66 ocean waves,67
and weather warnings.68 Imagery can be processed for features and objects extracted
so that entities such as building footprints69 and roads can be mapped and counted.70
Many commercial services now offer detailed and dynamic services, such as vehicle
and facility 71 detection.72 For supply-chain management and cargo tracking, Automatic
Identification System (AIS)73 and IoT74 services exist.
Broad indices can be derived from the values of pixels, such as for vegetation,75 water,76
and ocean colour.77 Pixel content within imagery can be classified, such as for water, a
type of vegetation, or artificial surfaces, to create maps of land cover,78 land use,79 and
urban areas.
Change detection, a common processing task, takes sets of collocated imagery and
detects changes over time. This can be used to detect new settlements, settlement
growth, or settlement removal, for example. Changes in the condition of infrastructure
can be determined for access routes.
There are a variety of software tools and techniques in the public domain to assist in
data processing.80 Machine learning (ML) is proving an increasingly important area for
satellite applications81 due to its ability to process large imagery datasets more rapidly
with a variety of learning algorithms in combination with changes in the affordability
of highly parallel processing technology. Several initiatives exist to provide access to
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
necessary tools and training data, notably Radiant Earth Foundation.82 Most existing
suppliers are adopting ML techniques, and new private sector companies are emerging
to specifically address opportunities, e.g., Impact Observatory.83
Satellite data is rarely the sole source of information required to address a humanitarian
emergency. Therefore, consideration must be given to existing sources of national
and local data and expertise, including scientific and research institutions, statistical
authorities, and domain expertise from the public and private sectors. For example, crisis
mapping organisations like the Humanitarian OpenStreetMap Team and MapAction
62 NASA Jet Propulsion Laboratory, ‘Shuttle Radar Topography Mission,’ updated 9 October 2021, www2.jpl.nasa.gov/srtm/
63 JAXA EORC, ‘ALOS Global Digital Surface Model,’ www.eorc.jaxa.jp/ALOS/en/aw3d30/
64 ESA, ‘SMOS Data—Earth Online,’ https://earth.esa.int/eogateway/missions/smos/data
65 NASA Earthdata, ‘VIIRS I-Band 375 m Active Fire Data,’ updated 7 July 2021, https://earthdata.nasa.gov/earth-observation-data/near-real-time/
firms/viirs-i-band-active-fire-data
66 Institute of Arctic and Alpine Research University of Colorado, ‘The Flood Observatory,’ http://floodobservatory.colorado.edu
67 Copernicus Marine Services, ‘Global Ocean Waves Analysis and Forecast,’ https://resources.marine.copernicus.eu/?option=com_
csw&view=details&product_id=GLOBAL_ANALYSIS_FORECAST_WAV_001_027
68 Indian Meteorological Department, ‘Warnings,’ https://mausam.imd.gov.in/imd_latest/contents/subdivisionwise-warning.php
69 European Commission, ‘Global Human Settlement Layer,’ https://ghsl.jrc.ec.europa.eu
70 Geo-Referenced Infrastructure and Demographic Data for Development, ‘GRID3 Data Hub,’ data.grid3.org
71 Aventior, ‘Insights from Aerial and Satellite Images,’ aventior.com/aerial-satellite-image-analysis/
72 4 Earth Intelligence, ‘Object and Vehicle Detection,’ www.4earthintelligence.com/products/object-and-vehicle-detection/
73 exactEarth, ‘exactAIS,’ www.exactearth.com/product-exactais
74 Hiber, ‘IoT Solutions,’ https://hiber.global
75 USGS, ‘NDVI, the Foundation for Remote Sensing Phenology,’ www.usgs.gov/core-science-systems/eros/phenology/science/ndvi-foundation-
remote-sensing-phenology?qt-science_center_objects=0#qt-science_center_objects
76 European Commission, ‘Global Surface Water Explorer,’ https://global-surface-water.appspot.com
77 NASA Goddard Space Flight Centre, ‘Ocean Color Web,’ https://oceancolor.gsfc.nasa.gov
50
78 ESA Climate Change Initiative, ‘Landcover,’ www.esa-landcover-cci.org/?q=node/164
79 OpenStreetMap, ‘Landuse Landcover,’ https://osmlanduse.org
80 ESA, ‘Sentinel Application Platform,’ http://step.esa.int/main/toolboxes/snap/; QGIS, ‘A Free and Open Source Geographic Information System,’
https://www.qgis.org/en/site/; OSGeo, ‘The Open Source Geospatial Foundation,’ www.osgeo.org
81 David J. Lary et al., ‘Machine Learning Applications for Earth Observation,’ in Earth Observation Science and Open Innovation, ed. Pierre-Phillippe
Mathieu and Christoph Aubrecht, 165–218 (Cham: Springer Open, 2018), https://doi.org/10.1007/978-3-319-65633-5_8.
82 Radiant Earth Foundation, ‘Who We Are,’ www.radiant.earth/about/
83 Impact Observatory, ‘About,’ www.impactobservatory.com/about
rely on international and local volunteers to ensure that the tracing of geographic
features in satellite imagery is correct and acceptable to affected communities.
Cloud services are a prominent model in the supply of satellite data by either cloud
suppliers or satellite operators. Cloud services with Infrastructure as a Service (IaaS)
make large imagery archives available within a cloud environment (storage) and provide
the IT infrastructure to run and host applications (computing). They also provide
a wide variety of supporting software tools, such as popular software development
libraries and environments, processing tools and algorithms, ML tools, and databases
and hardware (e.g., AI accelerators and sensors). Cloud services can provide quick
access to infrastructure that can be complex to independently develop and maintain
but still requires skills and expertise. Many cloud services make satellite imagery
available alongside tool sets as a standard service; examples include AWS,90 Planetary
Computer,91 and IBM Agriculture.92 What sets this model apart from “archive
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
Cloud providers also provide integrated ground stations, satellite operations, and data
handling services, including AWS Ground Station93 and Azure Orbital.94
Platforms provide a more specific and ready-to-use set of online tools and data that
often run on top of cloud services. Examples include Descartes Labs95 and Sentinel
Hub.96 Platforms can bring together mapping and geospatial analysis tools with online
imagery access. These include services from GIS suppliers and online mapping/virtual
globe providers, such as commercial suppliers ESRI,97 MapBox,98 Google,99 and
OpenStreetMap (OSM).100
84 United States Agency for International Development, ‘Famine Early Warning System,’ https://fews.net
85 European Commission’s Copernicus Emergency Management Service, ‘Global Flood Awareness System,’ www.globalfloods.eu
86 British Geological Survey, ‘METEOR,’ www.bgs.ac.uk/geology-projects/geodesy/meteor/
87 Copernicus, ‘Copernicus Emergency Management Service,’ https://emergency.copernicus.eu
88 Committee on Earth Observation Satellites, ‘Recovery Observatory,’ https://ceos.org/ourwork/workinggroups/disasters/recovery-observatory/
89 Mostapha Mohammad Harb and Fabio Dell’Acqua, ‘Remote Sensing in Multirisk Assessment: Improving Disaster Preparedness,’ IEEE Geoscience
and Remote Sensing Magazine 5, no. 1 (March 2017): 53–65 https://doi.org/10.1109/MGRS.2016.2625100
90 Amazon Web Services, ‘Registry of Open Data on AWS,’ egistry.opendata.aws/tag/satellite-imagery/
91 Microsoft, ‘Planetary Computer,’ planetarycomputer.microsoft.com
92 IBM, ‘IBM Agriculture,’ www.ibm.com/products/agriculture
93 Amazon Web Services, ‘AWS Ground Station,’ https://aws.amazon.com/ground-station/
51 94 Microsoft, ‘Azure Orbital,’ https://azure.microsoft.com/en-us/services/orbital/
95 Descartes Labs, ‘Products,’ www.descarteslabs.com/products/
96 Sinergise, ‘Sentinel Hub,’ www.sentinel-hub.com
97 Esri, ‘ArcGIS Image for ArcGIS Online,’ www.esri.com/en-us/arcgis/products/arcgis-online-imagery/overview
98 Mapbox, ‘Mapbox Satellite,’ www.mapbox.com/maps/satellite/
99 Google, ‘Google Earth Engine,’ https://earthengine.google.com
100 A global crowdsourcing and mapping project similar to open knowledge movements like Wikipedia. See www.openstreetmap.org.
101 Humanitarian OpenStreetMap Team, ‘Our Work,’ www.hotosm.org/our-work
crises that get major international attention. The UN’s Global Pulse initiative also uses
big data and ML for crisis insights.102
Google’s Earth Engine “is free for research, education, and nonprofit use” and the
Amazon Sustainability Data Initiative (ASDI) “seeks to accelerate sustainability research
and innovation by minimizing the cost and time required to acquire and analyze large
sustainability datasets.” 103
PNT signals are broadcast from a GNSS from space. These signals are augmented
to improve accuracy by ground stations for applications such as safety-critical aircraft
systems.105 Many operational GNSS satellite constellations now exist and are operated
by national governments.106 These include GPS (USA), BeiDou (China), Galileo (EU),
GLONASS (Russia), NavIC (India), and QZSS (Japan).
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
Whilst GNSS signals can be received globally with relative ease, augmentation
systems using ground-based infrastructure are needed for specific use cases such as
civil aviation. As augmentation systems are not yet available in some regions, the full
benefits of GNSS are not yet uniformly deployed globally.107 The EU, for example,
is cooperating 108 with African countries109 in the deployment of Satellite Based
Augmentation Systems (SBAS) for aircraft which reduce position errors for safer
landing, fuel-saving route optimisation, and airport equipment cost reduction.
GNSS receivers are found in a wide variety of devices; examples include mobile phones,
farming machinery, industrial assets, and navigation systems for ships and aircraft.
PNT signals are received by these receivers to establish accurate time and position for
the object. To describe “where things are,” GNSS positioning can be used together
with local reference systems (e.g., national mapping grids and street names) to record an
object’s location both globally and locally.
The combination of frequently updated location data for multiple assets, entities, and
objects (e.g., emergency vehicles or rescue team members in search and rescue missions)
together with timing and other datasets gives rise to location intelligence and analysis.
Also, geo-tagging relevant issues in the humanitarian context (e.g., infrastructure,
disasters, damages, conflict incidents, response activities, etc.) on the ground via GNSS
technology provides detailed contextual and positioning data to humanitarian actors.
Free and commercial platforms aimed at humanitarian and development work are
broadly available to assist in the collection, analysis, and management of location data.113
The Sharing the Land project at the Université Chrétienne Bilingue du Congo uses
GNSS and GIS to map neighbourhood conflicts and land claims to assist in dispute
resolution. In improving land administration, the project seeks to address land conflict
in eastern DRC with government and local community members and “to improve
land use planning by introducing a citywide strategic planning approach that takes into
consideration the informal growth of cities in peri-urban contexts.” 118
“A GPS unit (or maybe a satellite phone with built-in GPS) is a very useful extra item,
if you have one or can get access to one.”119
111 HARMLESS Consortium, Critical Analysis Report, European GNSS Supervisory Authority, 2007, www.euspa.europa.eu/simplecount_pdf/
tracker?file=virtual_library/2007-04-19_Critical_Analysis_Report_for_the_HARMLESS_Project.pdf
112 SPOT, ‘About SPOT,’ www.findmespot.com/en-us/about-spot/company-info
113 Eyad Ghattasheh, ‘Managing Syrian Refugee Camps Using ArcGIS,’ Esri, Fall 2017, www.esri.com/about/newsroom/arcuser/managing-syrian-
refugee-camps-using-arcgis/
53
114 Personal communication to Caribou Space, 2021
115 Harvard Humanitarian Initiative, ‘KoBoToolbox,’ www.kobotoolbox.org
116 Ushahidi, ‘The Ushahidi Platform,’ www.ushahidi.com/features
117 Tom Esslemont, ‘U.S. Incorrectly Mapped Afghan Health Facilities Twice—Watchdog,’ Reuters, 24 August 2015,
www.reuters.com/article/usaid-afghanistan-gps-idINKCN0QT25120150824
118 Integrated Research Institute IRI-UCB, ‘Sharing the Land,’ https://iriucbc.org/current-projects/sharing-the-land/
119 MapAction, Field Guide to Humanitarian Mapping, 2nd ed., July 2011, https://mapaction.org/wp-content/uploads/2016/12/mapaction_field_guide_
to_humanitarian_mapping_low_res.pdf
Custom satellite applications
Mobile apps for specific use cases are increasingly available from global app stores
or sideloading sources. Safecity, for example, facilitates anonymous crowdsourced
incident reporting of sexual violence which can reach epidemic proportions during mass
demonstrations and complex emergencies.120 These datasets assist stakeholder groups
and individuals in identifying high-risk areas, and additional apps facilitate alerting and
requesting assistance.121
Voice and data/internet services are now commonly available from both geostationary
satellites and numerous satellites in low earth orbit (LEO) constellations, e.g., OneWeb124
and Starlink.125 Common SatComms-based services include telephony, mobile satellite
services (MSS), and internet connectivity with all its associated digital services.
Comms and data equipment are required to use SatComms services and include very-
small-aperture terminals (VSAT) to provide reliable voice and data connections. These
can be deployed for permanent use at a single location, but most can be easily relocated.
Small data modems provide device and computer connectivity at lower speeds, and widely
available satellite phones (handsets) often include messaging and other service options.
For emergencies, many satellite operators have signed the Crisis Connectivity Charter.
Led by the WFP and the Emergency Telecommunications Cluster (ETC),126 operators
are “committing satellite equipment and capacity that will be dedicated for humanitarian
purposes during emergency responses.” 127 The International Telecommunication Union
(ITU), a UN agency, supplies satellite phones, terminals, and telecommunications
equipment upon request to its 193 Member States during a disaster response.128
120 Human Rights Watch, ‘Egypt: Epidemic of Sexual Violence,’ 3 July 2013, www.hrw.org/news/2013/07/03/egypt-epidemic-sexual-violence
121 bSafe, ‘Feature,’ www.getbsafe.com/features; Circle of 6, ‘Home,’ www.circleof6app.com
122 Inmarsat, ‘Satellites,’ www.inmarsat.com/en/about/technology/satellites.html
54
123 Intelsat, ‘Global Network,’ www.intelsat.com/global-network/
124 OneWeb, ‘Network,’ https://oneweb.net/network
125 SpaceX, ‘Starlink,’ www.starlink.com
126 UN OCHA Inter-Agency Standing Committee, ‘Emergency Telecommunications Cluster,’ www.etcluster.org
127 EMEA Satellite Operators Association, ‘Satellite Industry and UN to Enter Operational Phase of Crisis Connectivity Charter,’ 17 May 2018,
https://esoa.net/news/satellite-industry-and-united-nations-to-enter-operational-phase-of-crisis-connectivity-charter-for-support-of-global-
disaster-relief/
128 International Telecommunications Union, ‘ITU Disaster Response,’ www.itu.int/en/ITU-D/Emergency-Telecommunications/Pages/Response.aspx
Smart modules and devices (IoT)
Smart modules are installed in small devices, IT equipment, and machinery to connect
to networks of devices in the wider Internet of Things (IoT). Communication satellites
can receive transmissions from and connect to these devices; remote IoT devices can
connect to the internet directly via satellite. Connected sensors include water-level
gauges, air pollution sensors, container-borne GPS trackers, and temperature/humidity
sensors in food storage.
55
129 World Health Organization, Telemedicine: Opportunities and Developments in Member States, Report on the Second Global Survey on eHealth, 2009
(Geneva: WHO, 2010), https://apps.who.int/iris/bitstream/handle/10665/44497/9789241564144_eng.pdf ?sequence=1&isAllowed=y
Landscape of
humanitarian
satellite applications
Credit: ESA
Landscape of humanitarian
satellite applications
Key points
• 42% of identified humanitarian satellite applications targeted government
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
users.
• The highest number (92) of identified applications are for “food, security,
nutrition and famine” events in Africa.
The criteria for including a satellite application in the analysis are shown in Figure 13.
Additional details of the information collected are provided in Annex 3: Methodology
and limitations.
Satellite technology
Domain Use case Hazard
is core part
in scope in scope in scope
of solution
57
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Geography
Satellite applications for humanitarian emergencies are delivered globally. The dataset’s
geographical concentrations are mapped in Figure 14.
Kenya (24) is the country with the highest number of applications recorded, followed by
Bangladesh (21), Myanmar (16), Vietnam (16), India (15), and China (14).130
Whilst no countries have been explicitly excluded, the dataset was assembled with a
preference for applications in developing regions. It is also recognised that the search
was conducted in English, which may omit a significant percentage of non-English-
described applications.
Some satellite applications are present in multiple countries and are tagged “multiple,”
but not included in the map in Figure 14. Of the 127 of 248 countries with no
applications recorded, some are known to be addressed by this subset of applications.
This subset contains a variety of domains and use cases, but those related to “health
emergencies” are more prominent than in the whole dataset.
130 Kenya hosts two regional centres (RCMRD and ICPAC) that lead many EO-based regional projects. Kenya also hosts regional offices for the UN.
2.0%
3.1%
4.8%
8.9%
34.7%
16.9%
Asia
Africa
Multiple
LAC
29.6% Europe
Oceania
Northern America
The region with the most occurrences of applications is Asia, followed by Africa and
applications that address multiple regions.
The lowest number of applications found was in Northern America. Northern America
is understood to have a high use of satellite applications as one of the world’s most
developed regions; its low percentage in the dataset is considered the result of its
economic status and lack of emergency-related applications termed “humanitarian”
rather than, for example, public services. This distinction is not applied methodically in
all regions.
Event type
Different regions have varying concentrations of satellite applications for event clusters,
as shown in Figure 16. Specific events might have a higher or lower frequency than
expected, e.g., cyclones, as explained in Annex 3: Methodology and limitations.
2 Multiple Event Types (91), occurring mostly across multiple regions. These 91
applications are mostly EO (56%), but have a larger percentage of SatComms (27%)
compared to the dataset as a whole (6.6%), highlighting the general applicability of
SatComms to multiple event clusters and use cases. Within the EO applications, many
are used for more strategic/generic purposes, such as stakeholder coordination, damage
assessment, or emergency response planning. Applications that cover Multiple Event
Types are relevant for use in analysis of compound risks and in complex emergencies
(see section Compound risks and complex emergencies).
Flood
Refugees/IDPs
Illegal activity
Water supply
Cyclone
Wildfire
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
Insect Infestation
Complex emergencies
Invasive species
Oceania
Temperature-related (Freeze, Heatwave)
Latin America
Wind-related (Storm, Tornado) and the Caribbean
Africa
Lithometeors (Dust, Smoke)
Asia
Civil unrest Multiple
0 10 20 30 40 50 60 70 80 90 100
Figure 16: Frequency of Satellite Applications by Event Cluster and Region (n =>500)
There are a variety of event clusters less represented in the dataset, which may indicate
underserved areas requiring further investigation. However, these are hazard types that
are more common with greater sections of the global population and economy affected,
e.g., floods.
Supplier types
Applications come predominantly from Private Suppliers, followed by NGOs, Public
Suppliers, and Academia. There is little evidence of direct provision by Media and
Development Agencies.
As Private Suppliers is a large and diverse group, this is further subdivided. Supply
by Platform/Solution Providers is the most common, followed by Satellite Operator/
Resellers, Hardware/Software Suppliers, Cloud Computing Providers, and Other.
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
4.4%
%
4.8%
1.5%
5.5%
42.2%
Government
NGOs
9.0% Private Sector
Development agency
First responder
9.9% Affected public
General public
16.3% Academia
Media
Unknown
11.8%
40.0%
20.5%
Private supplier
61 NGOs
Public supplier
Academia
24.5% Media
Development agency — 0.6%
Unknown — 0.6%
Satellite Operator/Resellers supply 25.9% of applications, highlighting extensive
vertical integration as those satellite operators move downstream, not only selling
the data streams but also developing specific applications. Conversely, as 51.2% are
Platform/Solution Providers, which are downstream-only players, they need to access
the data streams from potentially competitive upstream Satellite Operator/Resellers.
Satellite applications may service more than one type of user; Figure 20 highlights
the relationships in terms of provision of satellite applications from the supply-side to
demand-side—which is different to the count of suppliers shown in Figure 18 and 19.
Figure 19: Proportion of Satellite Applications by Private Supplier Subtype (%) (n =>500)
2.0%
3.5%
17.4%
51.2%
Figure 20: Proportion of Satellite Applications in Terms of Relationship between Supplier Type and User Type (Supplier Count ≥500, Relationship Count ≥850)
NGOs 16%
22% NGOs
Private Sector 9%
62
First Responder 6%
11% Academia
Affected Public 5%
2% Media
General Public 4%
1% Unknown Academia 3%
Development Media 3%
0% Agency 1%
Unknown
Domains and events
As detailed previously, for the purposes of this report, humanitarian events are
classified according to five domains. Figure 21 shows the percentage of satellite
applications per domain.
Satellite applications occur mostly within the disasters (36.3%) domain, followed by
food insecurity (18.7%) and security and conflict (18.7%). Population displacement
(6.8%) and health emergencies (8.4%) contain the least number of applications.
11.2% of satellite applications address all domains. Of those, the vast majority also
address multiple event clusters, indicating a cross-cutting capability across domains,
events, and use cases. There are specific use cases that support all domains, for example:
infrastructure mapping and exposure, early warning systems, population and poverty
mapping, humanitarian stakeholder coordination, and connectivity for public or
remote populations. Also, from a technological point of view, SatComms applications
commonly cut across all domains and event types.
Use cases
The range of addressable problems for satellite applications is broad and diverse,
as shown by the variety of use cases detailed earlier. The use cases with the most
applications found are early warning systems (128), emergency response (60), and
damage assessment (60).
6.8% 11.2%
8.4%
18.7%
36.3%
All
Disasters
Food insecurity
18.7%
Security and conflict
Health emergencies
Population displacement
63
Figure 22: Frequency of Satellite Applications by Use Case (n =>500)
Early warning systems
128
Emergency response
60
Damage assessment
60
Crop yield and productivity forecasting
57
Hazardous event detection and severity
53
Population and poverty
52
Natural resource mapping
46
Infrastructure mapping and exposure
44
Humanitarian stakeholder coordination
33
Microplanning for health interventions
29
Refugee camps
28
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
The most common business model is a customised solution where the user pays for a
customised/bespoke application from a supplier. Open access applications are available
in the public domain to use free of cost and other access barriers. In the subscription
model, a user pays an ongoing fee for access to the service, but doesn’t own it. In research,
the satellite application has been developed primarily for research or academic purposes.
Finally, stakeholders can purchase “off-the-shelf ” and own the application from a supplier
with minimal customisation required.
11.2%
13.0% 1.6%
12.8% 61.5%
Customised solution
Open access
Subscription
Research
Off-the-shelf
65
Scale
In terms of geographies and numbers of users, satellite applications have scaled on
a spectrum, as highlighted in Figure 24. Each scale corresponds to an Application
Readiness Level (ARL). The following options are used for ‘Scale’:
• Implemented applications are operational services with limited geographic reach and
users (ARL 7–8).
• Scaled applications have sustained, long-term use across wide geographies, multiple
countries, and with multiple users (ARL 9). Evidence of repurposing a ‘successful’ pilot
or cross-applicability across multiple domains/events/use cases is also a contributor to
selection of this option, but is often not visible in the available literature.
The most common scale for satellite applications is implemented. A factor in the high
proportion of Implemented projects is the many programmes that provide funding to
supply-side players for projects that are usually designed to deliver to specific users in
specific countries. These initiatives focus on supporting ‘pilot-scale implementation’ but
are not intended to offer long-term financing for resulting applications. This topic is
explored further in section Financing for application development and scaling.
16.8%
26.7% 56.6%
66 Implemented
Scaled
R&D
In the top-left of Figure 25 are use cases, including humanitarian supply chain
management, which have significant data gaps and only eight satellite applications.
Supply chain management, like many other use cases, realises indirect value from
satellite imagery but necessitates other data sources like supply inventories, warehouse
locations, needs inventories, and more. Other use cases directly benefit from satellite
data (e.g., classifying damaged buildings from imagery) and do not require additional
datasets. The bottom-right shows use cases, like early warning systems or crop yield and
productivity forecasting, that have less extensive data gaps and many more applications
currently in use.
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
Figure 25: Frequency of Satellite Applications by HDX Data Gaps for Each Use Case (n ≥500) Most direct value from satellite data
Indirect value from satellite data
100 High demand/low supply High demand/high supply
Data gaps (%)
Search and
rescue missions Infrastructure mapping
and exposure
90
Humanitarian supply Emergency
chain management response
80 Damage
assessment
Contact tracing
Infrastructure
70 Disease prevalence reconstruction
Refugee
resettlement and transmission
suitability Refugee camps
Guiding drones Contaminated
sites Microplanning for
Subsidence
health interventions Hazardous event
and erosion
60 Monitoring detection and severity
Access routes and evaluation Extraction and trafficking of
and evacuation illegal resources or humans
plans
Nuclear facilities
Disaster risk
financing and insurance
Compliance with
peace treaties
Population
40 and poverty
Connectivity for public
or remote populations
67
10
Monitoring Evidencing
prison camps conflict-related violence
Key points
• Thirteen barriers to the uptake of satellite applications in humanitarian
emergencies are identified.
• Based on the ability of the development community to help address the issue,
a subset of eight barriers are prioritised for interventions:
− Data availability
− Procurement challenges
Barriers to uptake
This section identifies the barriers to uptake of satellite applications for humanitarian
emergencies and describes each barrier in detail. It also states whether addressing each
barrier is a priority for a development community intervention. Figure 26 identifies the
main barriers to increasing the uptake of satellite applications in humanitarian emergencies.
69 User awareness & resistance Inadequate monitoring & evaluation Inadequate understanding of needs Data availability
Regulatory constraints Ethics & privacy Technical expertise & skills Citizen safety
Across the demand-side, within the humanitarian community there are specific
organisations that champion the use of satellite applications, for example, UNOSAT
within the UN system. Equally, inside organisations, there are often individual
advocates who trial and test such innovations. However, there is significant variation
across the user community in the level of awareness of satellite applications, their use
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Cost structures can be unfamiliar and confusing (see section Procurement challenges).
Supply-side players are often unknown to members of the humanitarian community,
and it can be difficult to understand the technical capabilities within the supply-side
(i.e., the “art of the possible”) in order to appropriately articulate the user requirements
that need to be met. Thus, satellite applications might never be fully explored as an
option for the humanitarian community, or inappropriate or over-specified technical
solutions may be developed, which are then not valued by their users (see section
Inadequate understanding of needs).
Introduction of new technology, including satellite applications, carries the potential for
human and political resistance, as such technologies change—sometimes radically—the
methods for achieving a goal. For example, if national crop yield forecasts are produced
within the Ministry of Agriculture, the national government controls them. The same
forecast produced via a satellite application provided by a national or even international
Private or Public Supplier would mean a loss of control over such statistics. Many
people still fear job loss due to emerging technologies and new methods replacing
them. If they do not have experience with satellite data, these technologies can be
intimidating and involve steep learning curves in terms of technical specifications,
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
Trust in traditional methods has developed over many decades or centuries, and a
similar trust needs to be built with any replacement satellite-based method. The issue
of trust is often compounded by the so-called Black Box Problem with ML-derived
applications, whereby “a black box is a system which can be viewed in terms of its inputs
and outputs, without any knowledge of its internal workings.” 132 There are some efforts
tackling these transparency and replicability issues in terms of sharing guidance on
metadata standards, advocating for open source licensing, adopting frameworks based
on international human rights law and ethics, and more.
There are often numerous satellite applications that serve a similar purpose, for
example, global land cover. This makes it difficult for users to filter through the options
and know which is most appropriate for their context. Whilst there is strong supply-
side and demand-side expertise, what is lacking is ‘connector/bridge’ organisations and
persons that can link the two sides.
70
Sharing knowledge and raising awareness of the opportunities, use cases, existing
examples, and outcomes and impacts of satellite applications is a public good that is
appropriate for development community interventions.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
As identified in Table 3, with the use case Monitoring and evaluation, satellite
applications can directly support the M&E process. This includes “Geospatial Impact
Evaluations (GIEs) that rigorously evaluate the impacts and cost-effectiveness of specific
development interventions and large investment portfolios with spatial data. GIEs methods
leverage readily available data, like satellite observations or household surveys, to establish
a reliable counterfactual for measuring impacts—at a fraction of the time and cost of a
‘traditional’ randomized controlled trial (RCT).” 138 However, such techniques and impact
evaluations are currently rare.
There are existing initiatives upskilling the satellite applications sector on M&E
techniques. For example, UKSA IPP has embedded robust M&E throughout the
programme design and has trained, via Caribou Space, ~50 organisations on M&E.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
However, IPP worked with only a subset of organisations from the global sector. This is
a long-term barrier which will require continued, extensive effort from multiple parties
to address.
Often, there is a misalignment between the technical solution and the nature of the
problem (user requirements). Supply-side organisations often don’t fully understand
the intricacies of the humanitarian domains, and if demand-side organisations are
not available or able to communicate specific user requirements, misalignment occurs.
This commonly manifests in over-engineering, over-customisation, and/or over-
simplification of the satellite application, leading to lengthy delivery time frames, high
costs, and complex operations/maintenance.
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
A related issue is “over-selling” the capabilities of satellite applications, which can lead
to inaccurate and non-validated information. This can then lead to disappointment
and mistrust in user communities when results are not achieved or even relevant for a
specific stakeholder.
Much more interaction, especially between Affected Communities and data producers,
72
is needed to better determine the appropriate use of data and technology.
139 Jos Berens et al., ‘The Humanitarian Data Ecosystem: The Case for Collective Responsibility,’ Stanford Center on Philanthropy and Civil Society,
2017, https://pacscenter.stanford.edu/wp-content/uploads/2017/11/humanitarian_data_ecosystem.pdf
140 Caribou Space, ‘Beyond Borders. Community Engagement Workshop 3,’ 2021
box 3: Overcoming Inadequate Understanding of Needs
This barrier is common when any form of technology is applied to humanitarian contexts,
including, for example, mobile or other ICT-based applications. The importance of
understanding user needs is embedded into the Principles for Digital Development.141
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Under these principles, donor organisations are expected to place a high priority on the
supply-side organisations to ensure robust understanding of needs via grant-funded
deliverables such as Political Economy Analysis, User Requirements Documents, and
Long-Term Sustainability Plans. Understanding user needs should continue to be a
principle for any donor-funded programme.
Given the ongoing existing initiatives to address this barrier, we do not identify it as a
priority barrier for a new development community intervention.
Data availability
addition, accessing data in a timely fashion may be affected by other demands and
tasking requests on the available satellites, meaning that humanitarian needs may not
be met as rapidly as required.
Within the demand-side, a lack of coordination between user organisations, and even
within single organisations, can lead to a lack of awareness of who owns or has procured
satellite applications or data. Organisational silos are common for both supply-side and
demand-side organisations, equally affecting private and public organisations. These
silos are “ business divisions that operate independently and avoid sharing information.” 142
This can lead to duplicative resource-intensive procurements of EO data, inefficient
costs, and limited usage. Further, the products are often inaccessible to local
communities. This is compounded by data-sharing agreements that often do not allow
sharing raw data.
For satellite applications to have value, they need significant data from terrestrial
sources to be combined. For example, in assessment of damaged buildings, satellite data
will not highlight how many people were inside the building. For flooding applications,
population datasets need to be combined to forecast how many people are affected by
the flood.
Figure 23 shows that the most common business model is a customised solution
where the user pays for a customised/bespoke application from a supplier (61.5%). This
highlights that there may be many duplicative satellite applications being developed for
separate users but with very similar requirements.
73
Part of the cause of this barrier is a lack of visibility of who is doing what and where,
leading to coordination failures and duplication of effort. This is a classic issue across
the development community and equally in satellite applications. Addressing it
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Examples of potential interventions would include open and free information sources
of who is doing what and where on the topic of satellite applications for humanitarian
contexts. An example of such an information source is an online open knowledge
base, and virtual or real-world knowledge events (same as for User awareness and
resistance barrier).
Regulatory constraints
Regulation and legislation can introduce significant operational and legal barriers for
the use of satellite applications. Conversely, an absence of regulation or enforcement
is potentially undesirable; for example, unregulated spectrum use can lead to radio
interference in satellite signals.143
There are five UN Committee on the Peaceful Uses of Outer Space (COPUOS)
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
A recent example of national law reducing barriers is India’s February 2021 geospatial
guidelines, which removed the need for licences and approvals for Indian organisations
to create and publish geospatial data (with sensitive exceptions).147 Indian entities are
now allowed to access ground stations and augmentation services for more accurate
GNSS real-time positioning and other previously restricted public survey datasets.
74
143 Jonathan Amos, ‘Europe’s Smos ‘Water Mission’ Battles Interference,’ BBC News, 5 May 2010, http://news.bbc.co.uk/1/hi/sci/tech/8661228.stm
144 UN Committee on the Peaceful Use of Outer Space Legal Subcommittee, 60th Sess., Status and Application of the Five United Nations Treaties
on Outer Space, UN Doc A/AC.105/C.2/2021/CRP.10 (31 May 2021), www.unoosa.org/res/oosadoc/data/documents/2021/aac_105c_22021crp/
aac_105c_22021crp_10_0_html/AC105_C2_2021_CRP10E.pdf
145 UN Office for Outer Space Affairs, ‘World Space Agencies,’ www.unoosa.org/oosa/en/ourwork/space-agencies.html
146 UN Office for Outer Space Affairs, ‘National Space Law,’ www.unoosa.org/oosa/en/ourwork/spacelaw/nationalspacelaw/index.html
147 Government of India, ‘Guidelines for acquiring and producing Geospatial Data and Geospatial Data Services including Maps,’ 15 February 2021,
https://dst.gov.in/sites/default/files/Final%20Approved%20Guidelines%20on%20Geospatial%20Data.pdf
These processes face considerable evolving challenges, including:
• Increasing collision and debris risk from satellites, constellations, and mega-
constellations in LEO
• Political challenges149
There are many standards relevant to the implementation and delivery of humanitarian
solutions (e.g., OGC,150 EU INSPIRE,151 UN-GGIM152) that address interoperability
and common mapping representations, for example. Absence of such standards
represents a barrier to uptake, transferability, and integration into wider solutions.
However, standards can be complex and difficult to implement, and may also represent
a complexity and cost barrier for humanitarian organisations.
Humanitarian organisations must also be aware of export controls, which are used
by countries “to protect national security interests and promote foreign policy objectives.” 153
Governments control the export of sensitive equipment, software, and technology for
reasons of national security and foreign policy; these regulations cover SatComms,
GNSS, and EO. Suppliers assume responsibility for compliance; for every potential
customer, they must know the intended use, end user contact details, and ultimate
destination of the data.154 Supply-side providers cannot distribute export-controlled
materials to individuals or companies in “embargoed countries.” 155 Commercial and
export controls make states having access to “available analysed information concerning
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
the territory under its jurisdiction” difficult to implement universally across borders.
When humanitarian emergencies occur in “embargoed countries,” it is not possible
to purchase from supply-side companies based in supplier countries with such export
controls. Organisations based in those embargoed countries are also restricted from
purchasing data from these supply-side companies due to such government-specific
regulations.
Besides specific regulations, many national governments have policies in place to direct
spending to domestic businesses; as a result, the countries that spend the most on space
programmes are home to the largest suppliers. There is an intentional bias to keeping
that spending in-country, and this directly influences countries’ technical capacity to
use satellite data.
148 Peter B. de Selding, ‘Signal Interference Proposal Could Make the ICT a Watchdog with Some Teeth,’ SpaceNews, 10 October 2014,
https://spacenews.com/42147signal-interference-proposal-could-make-the-itu-a-watchdog-with-some/
149 Elizabeth Hoffman and Kristen Cordell, ‘An Obscure UN Agency Guides Digital Communications—Congress Must Endorse US Leadership',
The Hill, 23 July 2021, https://thehill.com/opinion/international/564527-an-obscure-un-agency-guides-digital-communications-congress-must
75
150 Open Geospatial Consortium, ‘About OGC,’ www.ogc.org/about
151 European Commission, ‘INSPIRE Geoportal,’ https://inspire-geoportal.ec.europa.eu/theme_selection.html?view=qsTheme
152 UN Statistics Division, ‘Committee of Experts on Global Geospatial Information Management,’ https://ggim.un.org
153 International Trade Administration, ‘US Export Control,’ www.trade.gov/us-export-controls
154 Maxar, ‘Maxar Trade Compliance Policy,’ 2021, https://maxar-marketing.s3.amazonaws.com/files/legal/201112%20Maxar%20Trade%20
Compliance%20Policy%20-%20final%20EXTERNAL.pdf
155 United States Treasury, ‘Sanctions Programs and Country Information,’ https://home.treasury.gov/policy-issues/financial-sanctions/sanctions-
programs-and-country-information
box 5: Regulatory Constraints
These UN agencies are part of the development community, and their interventions
are required over the long term to address this barrier. However, we do not identify
this as a barrier for development community organisations outside of the UN system.
For humanitarians, anticipating potential risks in using satellite data can also be
difficult, as many of these issues are context-specific. For example, VHR EO can
produce detailed geo-referenced datasets over populations, including vulnerable
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
Informed consent is also impossible when using satellite data sources. The person or groups
involved may not be aware of this invasion of privacy or, if they are aware, are unlikely to
be able to mitigate it or easily take remedial action, in real time or retrospectively.158 Rupert
Allan, former country manager of HOT in Uganda, recommends that organisations think
about who is directing and controlling the collection of data, and to what degree it is
actually “accessible and accountable to and by local communities.” 159
Machine learning is a part of many EO-based satellite applications used to process the
data into information and insight. ML, both inside and outside of satellite applications,
has a range of ethical risks, including automated decision-making, inherent bias,
and misleading synthetic data.160 These risks can diminish trust in ML-produced
results and, given that much of the world’s ML talent resides in wealthier countries,
algorithms are biased to reflect those cultures and societal needs.
156 Matt Swayne, ‘Researchers Detail Privacy-Related Legal, Ethical Challenges with Satellite Data,’ Phys.Org, 12 July 2019,
76 https://phys.org/news/2019-07-privacy-related-legal-ethical-satellite.html
157 Günter W. Hein, ‘Location Privacy Challenges and Solutions, Part 1,’ Inside GNSS, 18 September 2017,
https://insidegnss.com/location-privacy-challenges-and-solutions/
158 G. Berman, S. de La Rosa, and T. Accone, Ethical Considerations when Using Geospatial Technologies for Evidence Generation, Innocenti Discussion Paper
2018-02, (Innocenti, Florence: UNICEF Office of Research, June 2018), www.unicef-irc.org/publications/pdf/DP%202018%2002.pdf
159 Malia Politzer, ‘Why We Need to Think about Ethics when Using Satellite Data for Development,’ Devex, 25 February 2021,
www.devex.com/news/why-we-need-to-think-about-ethics-when-using-satellite-data-for-development-99148
160 Cem Dilmegani, ‘AI Ethics: Top 9 Ethical Dilemmas of AI and How to Navigate Them,’ AI Multiple, updated 5 October 2021,
https://research.aimultiple.com/ai-ethics/
box 6: Overcoming Ethics and Privacy
Addressing ethics and privacy is not the natural focus of industry actors, and national
governments have varying standards and regulations regarding ethics and privacy. The
development community, with its focus on improving the lives of the world’s citizens,
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Specific skills are required by people in supply-side organisations (see Supply-side actors),
for example a software engineer in a Platform/Solution Provider or a GIS specialist in
an NGO, to produce satellite applications. This includes skills in EO, remote sensing,
GIS, software engineering, data science, ML, and co-development with domain-specific
expertise to ensure relevance and usefulness. All these skills are highly prized, and staff
turnover following training programmes is an issue for skills retention.
Skills are also required by the demand-side actors (see Demand-side actors), such as an
analyst in Government or a First Responder, to use satellite applications. This includes skills
in model development, application operation, data interpretation, combination with other
datasets, integration of ML approaches, and linkage of outputs to existing workflows.
However, there is also the need for very basic, simple-to-use tools that support
humanitarian emergencies. An example of this is Field Papers, which allows maps to be
printed and used rapidly in emergency contexts.163 Many of these tools are built as open
source by the open knowledge community.
There are many online learning resources available that support the development of
these skills, including tools, platforms, workshops, and tutorials. For example, Radiant
Earth’s ML Hub provides such resources specifically for machine learning for EO.164
77
161 Faine Greenwood, Caitlin Howarth, Danielle Escudero Poole, Nathaniel A. Raymond, and Daniel P. Scarnecchia, The Signal Code: A Human Rights
Approach to Information During Crisis, Harvard Humanitarian Initiative, January 2017, https://hhi.harvard.edu/publications/signal-code-human-
rights-approach-information-during-crisis
162 EthicalGEO, ‘Locus Charter,’ https://ethicalgeo.org/wp-content/uploads/2021/03/Locus_Charter_March21.pdf
163 Field Papers, ‘Welcome to Field Papers,’ http://fieldpapers.org
164 Radiant Earth Foundation, ‘Machine Learning for Earth Observation,’ www.radiant.earth/mlhub
box 7: Overcoming Technical Expertise and Skills
There are existing initiatives addressing the technical expertise and skills barrier,
including UKSA IPP via training of 7,200 persons and ESA GDA via Activity 3
Skills Transfer.165 There are extensive online training courses, both paid and free, for
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Also, there could be support to regional centres of expertise, such as the Regional
Centre For Mapping of Resources for Development (RCMRD)168 and/or IGAD
Climate Prediction and Applications Center (ICPAC)169 in Eastern Africa,
AGRHYMET170 in west Africa, and Asian Disaster Preparedness Center (ADCP)171
and/or ICIMOD172 in Asia.
Citizen safety
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
Powerful technologies often carry the risk of malevolent use. Satellites transmit
sometimes sensitive data to and from Earth, making them a potential target for
hackers or malicious actors. Satellite data can be doctored to spread misinformation, for
example, with erroneous positioning, deep fake imagery, or falsified communications
intercepts. In addition to intentional misinformation, there is also the potential for
well-intended analysts to incorrectly interpret an image or release information that
undermines the well-being of populations.173 As technical barriers are lowered, there
is increasing risk that SatComms can be hacked and transmitted information can
be leveraged by bad actors, affecting military command systems, launch systems,
communications, telemetry, and tracking.174 From a security perspective, the radio
frequency interference from electronic devices, radio antennas, or modems is strong
enough to drown out the relatively weak signals to GNSS receivers. That vulnerability
creates a tempting target for bad actors.175
165 Caribou Space, Adoption and Impact of Earth Observation for the 2030 Agenda for Sustainable Development, (Farnham, Surrey, United Kingdom.
Caribou Space, 2020), https://eo4society.esa.int/wp-content/uploads/2020/07/Caribou-Space_ESA-EO-for-Agenda-2030-v2.pdf
166 UK Space Agency, Space for Smarter Government Programme, ‘Training,’ https://spaceforsmartergovernment.uk/training
167 UNITAR, ‘UN Satellite Centre UNOSAT,’ https://unitar.org/event/event-pillars/united-nations-satellite-centre-unosat
168 Regional Centre for Mapping of Resources for Development, ‘About RCMRD,’ www.rcmrd.org/about-us/about-rcmrd
169 IGAD Climate Prediction and Applications Centre, ‘About Us,’ www.icpac.net/about-us/
170 UN Office for Outer Space Affairs, UN-SPIDER Knowledge Portal, ‘AGRHYMET Regional Centre,’
www.un-spider.org/agrhymet-regional-centre
171 Asian Disaster Preparedness Center, ‘About Us,’ https://www.adpc.net/igo/contents/adpcpage.asp?pid=2
78 172 International Centre for Integrated Mountain Development, ‘Who We Are,’ https://www.icimod.org/who-we-are/
173 Steven Livingston, ‘Satellite Imagery Augments Power and Responsibility of Human Rights Groups,’ Brookings Institution, 23 June 2016,
www.brookings.edu/blog/techtank/2016/06/23/satellite-imagery-augments-power-and-responsibility-of-human-rights-groups/; Stefan Voigt et al.,
‘Towards Semi-Automated Satellite Mapping for Humanitarian Situational Awareness,’ IEEE 2014 Global Humanitarian Technology Conference,
https://elib.dlr.de/95502/1/VoigtEtAl_IEEE_GHTC_06970315.pdf
174 Airbus, ‘Protecting Everyday Life—How Airbus Protects Satellite Systems Against Attacks,’ https://airbus-cyber-security.com/news/protecting-
everyday-life-how-airbus-protects-satellites-against-attacks/
175 Ingo Baumann, ‘GNSS Cybersecurity Threats: An International Law Perspective, Inside GNSS, 3 June 2019, https://insidegnss.com/gnss-
cybersecurity-threats-an-international-law-perspective/
Malevolent actors might use satellite applications directly against either other countries
or their own citizens, for example, the use of Satellite Sentinel Project’s imagery analysis
to guide combatants’ next attack on civilians.176 The Project’s goal was to acquire and
analyse imagery that would deter further aggression prior to the Sudan referendum. By
broadcasting perceived indicators of impending aggression, SSP inserted itself directly
into events and inadvertently provided valuable intelligence to bad actors, which resulted
in the kidnapping and ransoming of Chinese construction workers.177
Beyond Borders: Satellite Applications for Humanitarian Emergencies
IT infrastructure
Cloud platforms, such as Google Cloud Platform (GCP), Amazon Web Services
(AWS), and Microsoft Azure, address these challenges by storing and processing large
EO datasets in the cloud rather than on local IT infrastructure. They are significantly
more cost effective and have enabled large processing that was previously not possible
without sophisticated computing clusters. However, these platforms still require
subscriptions and robust internet connections to retrieve results. Google Earth Engine
(GEE) is a cloud platform that uses GCP and provides free access for users to conduct
analyses that would otherwise require enormous resources to access (download),
store, and analyse. This data and GEE’s cloud computing capabilities are available
to scientists, researchers, and developers, lowering the cost and time of obtaining,
pre-processing, and developing products from satellite imagery and building use-case
specific applications that ease data analysis and visualisation.
“Vendor lock-in” can be an issue, whereby supply-side organisations build their initial EO
applications on a specific cloud platform and then later face issues with switching to other
providers; for example, workflows built on AWS might not necessarily run on GCP.
Whilst cloud platforms reduce storage and processing requirements, they still require a
robust internet connection to access outputs. Internet access in developing regions is a
well-documented issue, so is not a major focus of this report. Barriers to internet access
include lack of mobile coverage, poor access to smartphones or desktop computers,
government taxes raising prices on bandwidth, and many more.178
79
176 Steven Livingston, ‘Satellite Imagery Augments Power and Responsibility of Human Rights Groups,’ Brookings Institution, 23 June 2016,
www.brookings.edu/blog/techtank/2016/06/23/satellite-imagery-augments-power-and-responsibility-of-human-rights-groups/
177 Michael Blanding, ‘Inside Harvard’s spy lab: How a team of Harvard geeks is using a satellite—plus a little help from George Clooney—to rewrite
the rules of humanitarianism,’ New York Times, 29 April 2012, www.bostonglobe.com/magazine/2012/04/28/inside-george-clooney-harvard-spy-lab/
RB6fK8MUYkBn3RvWFZpPqO/story.html
178 Alliance for Affordable Internet, ‘A4AI Policy and Regulatory Good Practices,’ https://a4ai.org/good-practices/; GSMA Connected Society,
Accelerating Mobile Internet: Policy Considerations to Bridge the Digital Divide in Low- and Middle-Income Countries, May 2021,
www.gsma.com/mobilefordevelopment/wp-content/uploads/2021/05/Accelerating-Mobile-Internet-Adoption-Policy-Considerations.pdf
Many countries have invested in National Spatial Data Infrastructures to efficiently manage
geographic data, metadata, users, and tools that are interactively connected.179 However,
the majority of these are in developed countries or multi-governmental institutions, such as
the European INSPIRE Initiative, UN Spatial Data Infrastructure (UNSDI), or World
Meteorological Organization (WMO) Regional Climate Centres (RCCs).
However, cloud platforms now significantly reduce the need for these at a national
level for each individual country. Cloud platforms provide three basic services: 1)
Beyond Borders: Satellite Applications for Humanitarian Emergencies
Individual satellite applications, and the supply-side organisations that produce them,
require financing. When these applications—or the core purpose of an organisation—
are considered to be of value to society (i.e., a public good), there may be a solid
rationale to support the provision of public, taxpayer-generated funding via grants.
Grants tend to be provided by governments and development agencies for a very
specific “public good” purpose. This purpose might include academic research, or work
to support sustainable development or humanitarian efforts. Typically, funding will be
required for short-term projects lasting one to three years, with reporting obligations
and M&E activities attached to encourage full accountability of the grant recipients.
For many NGOs, this project-based funding is an important contribution towards their
operating costs, with core “unallocated” funding much harder to attract.
Private companies seeking these grants may go from grant to grant, seeking funds
for specific projects, thus neglecting longer-term commercialisation prospects or
scalability. As a result, organisations reliant on publicly funded grants may have less
scope to innovate and to adopt a longer-term perspective, to the detriment of building
organisational capacity and expertise. Project funding may also be limited in scope,
with insufficient funds for needs assessments of on-the-ground organisations or to cover
salaries of highly technical staff with in-demand skills.
80
The landscaping work in this report has highlighted that there are a number of
initiatives that have provided financing to support industry to develop and pilot
applications in humanitarian settings, resulting in a high number of applications at
Beyond Borders: Satellite Applications for Humanitarian Emergencies
investment. Unlike grants, this capital will require repayment or an exchange of equity
and can be particularly difficult and/or expensive to attract. This type of financing is
typically even more difficult to find in developing regions where the existence of “venture”
and early-stage capital is limited. Here, the nascent satellite application sector may find it
difficult to access the capital they need to develop and then market their products.
Mission misalignment
The business model of Private Suppliers is focused on extracting the maximum value
from a product, so reselling an EO dataset multiple times is sensible. However, if the
user is a Government or NGO with multiple agencies, like the UN, they might be
able to reuse that dataset across many agencies. The private sector’s focus on profit also
leads to proprietary ownership in data processing algorithms, which again makes good
business sense, but potentially leads to concentrations of power within a few suppliers.
81
Satellite Operators have data licensing and ownership agreements with private and
national ground-receiving stations. There have been cases where receiving stations with
data rights have been slow to release imagery during humanitarian emergencies outside
the control of the satellite operator.182
processes of the suppliers and users. However, whilst these differences exist, Private
Suppliers are valuable members of the supply-side, and efforts should therefore be
placed on bridging the differences.
Even if supply-side and demand-side organisations use the same formal language,
huge differences in vocabulary, from scientific/technical (suppliers) to domain-specific
language (users), often lead to confusion. For example, complex disaster resilience
terminology may not be shared across organisations.
Organisations naturally and healthily have different missions and processes, so that
issue in itself is not directly addressable. However, it is possible for organisations
with differences in mission and processes to gain a better understanding of those
differences through ongoing engagement and collaboration.
Therefore, no specific development community interventions for this barrier are identified.
Procurement challenges
Procurement processes are ultimately the connecting and contractual link between the
supply-side and demand-side, and there are many challenges within them.
82
While many across the supply-side freely share technical specifications and
demonstrations, they rarely publish pricing. To understand which applications are
fit-for-purpose but also affordable, demand-side actors must engage with each provider
Beyond Borders: Satellite Applications for Humanitarian Emergencies
and/or manage a time- and resource-intensive Request for Proposals (RFP) process.
Also, understanding the supply-side’s licensing terms requires procurement and legal
teams who are knowledgeable on these topics that might not be available in house.186
Companies often cannot offer lower prices to the public than those offered to
government customers. Thus pricing is driven by their primary customer base: defence
and intelligence agencies. This outside purchasing power leads to unaffordable satellite
data and pricing for the nonprofits and startups in the demand-side, as well as licensing
terms tailored to those defence and intelligence customers.187
However, most Private Suppliers have discounts for humanitarian organisations and host
open data repositories where humanitarians can leverage released datasets for disaster
response or other noncommercial purposes. Companies like Planet188 and Maxar offer
free satellite imagery to universities for research purposes, as well as support disaster
response with public releases of before and after imagery over affected communities.189
partners in humanitarian response and offering more flexible data licensing terms for
those users.190 Along these lines, PLACE is a non-profit data trust that creates and
manages hyperlocal mapping data for a membership community that is open to all
and serves the public interest.191 In the forestry sector, the Norwegian Government’s
International Climate and Forests Initiative (NICFI) is sponsoring universal access to
high-resolution satellite monitoring of the tropical forests.192 The NICFI has awarded
a US$44M (£33M) contract to EO specialists Airbus, Planet, and Kongsberg Satellite
Services (KSAT). Andreas Dahl-Jørgensen, managing director of NICFI, said, “the
offered licensing terms are particularly good in this bid. It allows everyone to access high-
resolution satellite data, without restrictions on use and distribution.” 193
186 Caribou Space, ‘Beyond Borders. Community Engagement Workshop 2,’ 2021
187 Joe Morrison, ‘The Commercial Satellite Imagery Business Model is Broken,’ Medium, 7 August 2020, https://joemorrison.medium.com/the-
commercial-satellite-imagery-business-model-is-broken-6f0e437ec29d
188 Planet, ‘Disaster Data,’ www.planet.com/disasterdata/
83 189 Maxar, ‘Open Data Program,’ www.maxar.com/open-data
190 Amazon Web Services, ‘Solving Problems with Open Data Imagery: Q&A with DigitalGlobe and HOT,’ AWS Public Sector Blog, 1 February 2017,
https://aws.amazon.com/blogs/publicsector/solving-problems-with-open-data-imagery-qa-with-digitalglobe-and-hot/
191 Stefaan Verhulst et al., ‘Establishing a Data Trust: From Concept to Reality,’ PLACE,
www.thisisplace.org/blog-1/introducingplace/establishing-a-data-trust
192 Tara O’Shea, ‘Universal Access to Satellite Monitoring Paves The Way To Protect The World’s Tropical Forests,’ Planet, 2 March 2021,
www.planet.com/pulse/universal-access-to-satellite-monitoring-paves-the-way-to-protect-the-worlds-tropical-forests/
193 Norway’s International Climate and Forest Initiative, ‘New Satellite Images to Allow Anyone, Anywhere, to Monitor Tropical Deforestation,’ 23
September 2020, www.nicfi.no/current/new-satellite-images-to-allow-anyone-anywhere-to-monitor-tropical-deforestation/
box 12: Overcoming Procurement Challenges
Ensuring that procurement processes are simple and effective is the remit of supply-
side organisations. However, supporting demand-side organisations, e.g., Development
Agencies and NGOs, to be more intelligent customers isn’t always in the best interest of
Beyond Borders: Satellite Applications for Humanitarian Emergencies
supply-side organisations. Also, as detailed above for the forestry sector, bulk procurement
is a mechanism that could address this barrier.
Figure 23: Proportion of Satellite Applications by Business Model shows that the
most common business model is a customised solution where the user pays for a
customised/bespoke application from a supplier (61.5%). Also, Figure 24: Proportion of
Satellite Applications by Scale shows that most satellite applications are categorised as
implemented (56.6%), with only 26.7% categorised as scaled.
These statistics reflect a well-known barrier in the use of satellite applications for
humanitarian and broader development purposes: piloting and duplication. That is
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
where projects or products are implemented with a single user or a small group, often
in a single country, but do not scale beyond that initial implementation to become a
more broadly adopted application, often termed pilots. One-off, customised solutions or
pilots will struggle to satisfy global market demand.
Piloting and duplication is driven by multiple factors. First, members of the supply-side
often see the “project” as the final result, without planning or aiming for development
of a long-term sustainable offering.
Finally, the greater the need for local customisation of a solution, e.g., the need to
combine large volumes of local ancillary data, or for local languages, the more likely
this barrier will be an issue.
84
box 13: Overcoming Piloting and Duplication
Whilst piloting and duplication is natural, effort should be made where possible to avoid
unnecessary duplication. Multiple applications developed for the same and/or different
stakeholders can cause confusion for end users (e.g., deciding what application to use). It is
Beyond Borders: Satellite Applications for Humanitarian Emergencies
also important for donors to consider the long-term sustainability and scalability of their
investments and for financing to align well to operational needs beyond the shorter-term
testing of new ideas and innovations.
User awareness & resistance Inadequate monitoring & evaluation Inadequate understanding of needs Data availability
Regulatory constraints Ethics & privacy Technical expertise & skills Citizen safety
85
Table 4: Summary of Priority Barriers and Recommended Development Community Interventions
Data availability • Online open knowledge base of who is doing what and where
• Virtual and real-world knowledge events
• Donors to fund high-resolution EO data for use by humanitarian organisations
Ethics & privacy • Documenting and sharing ethics and privacy best practices
• Upscaling community involvement in data generation through community mapping and validation exercises
Technical expertise • A playbook of principles and case studies that showcase good practice
& skills • Support to regional centres of expertise
Financing for application • Appropriate patient capital from development agencies and philanthropic organisations through the
development & scaling creation of new financing mechanisms
Procurement challenges • Training for demand-side organisations on satellite application procurement processes and pricing
• Mechanisms to aggregate demand into a bulk procurement agreements across multiple demand-side
organisations
Piloting & duplication • Information of who is doing what and where via an online knowledge base
• Virtual and real-world knowledge events
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
86
Conclusions
Conclusions
This report describes the current landscape of satellite applications being used in
humanitarian emergencies and provides a ‘state of play’ snapshot of a rapidly evolving
sector. Satellite applications developed by an increasingly complex web of supply-side
stakeholders can be used across a wide range of use cases throughout the life cycle of an
emergency to improve our understanding of hazards, assessment of vulnerabilities, and
deployment of capabilities.
However, despite a large number of applications that have been piloted in the
humanitarian sector, there is a limited body of evidence to offer humanitarians
guidance on where satellite technology can be used most cost-effectively and with
the best outcomes for affected populations. Furthermore, rapid innovation and
increasing piloting has led to some duplication of effort on the supply-side, and the user
community has not yet fully acquired an understanding of the potential relevance for
Beyond Borders: Satellite Applications for Humanitarian Emergencies
their work and the capabilities of the technology. Regulatory and legal frameworks have
also struggled to keep pace with new developments.
There is now an opportunity for public and private sector stakeholders to reflect on
what more could be done to increase the use of, and impact derived from, satellite
applications in humanitarian assistance. Over the coming months, the Beyond Borders
project will be developing recommendations on what actions could be taken and the
potential outcomes for such interventions.
Conclusions
88
Annexes
Annex 1
Philippines Typhoon Haiyan case study
David Garcia, a team member, has documented a specific case study regarding
Philippines Typhoon Haiyan – a summary of which is below.
Before, during, and after the disaster, hundreds of mapping projects, exercises,
and research studies were undertaken to assist in the planning, response, and
reconstruction. Many of these projects were done in coordination with Filipino
mapmakers, emergency professionals, and internet users who dealt with both the social
and technical aspects of making humanitarian geospatial work useful for, and relevant
to, affected populations.
Annex 1 Beyond Borders: Satellite Applications for Humanitarian Emergencies
The case study provides an overview of how geospatial technologies, including those
related to satellites, EO, and remote sensing, can be useful in a major disaster like Super
Typhoon Haiyan. This includes the use of geohazard mapping and satellite systems,
crisis dashboards and government geoportals, and community, non-profit, and private
sector mapping. The case study also highlights three key issues in the production of
geographic information. These are detailed in the full case study to be published in
2022 and summarised here:
• Accuracy – The accuracy of the spatial data is important in terms of coverage, error,
damage, and other aspects of how features like buildings and roads were traced. In this
challenge, higher-resolution imagery is demanded to increase the quality of spatial data.
• Accessibility – Lack of access, complex software systems, and poor internet connections
impede the usability of the maps for the communities who directly respond to crises.
90
Annex 2
East Africa case study
Stella Chelangat Mutai, a team member, has documented a specific case study
regarding East Africa – a summary of which is below.
This case study analyses the use of satellite applications in the humanitarian context in EA.
It also seeks to understand stakeholder engagement and contributions by highlighting gaps
and determining possible entry points in the use of satellites in humanitarian decision-
making and policy-driven assessments within the identified countries. This research
Annex 2 Beyond Borders: Satellite Applications for Humanitarian Emergencies
• The level of capacity and knowledge within EA to identify modalities of training that
are needed
Climate Change, Land Cover, Natural Hazards, and Food Insecurity, respectively, were
voted as the potential use case domains in which increasing satellite technology capacity
would have great impact. This was influenced by the fact that most domains affect day-
to-day livelihoods of people within the case study region, and there is a dire need for
efficient early warning systems and actionable information for decision-making.
The research also sought to understand the level of awareness of existing satellite
applications and the technical specifications of satellite data within existing
humanitarian organisations. The survey found:
• 73% of humanitarian organisations within the scope of study have a broad knowledge of
91
existing data and use cases; however, 27% have limited to no information.
194 Rhoda Margesson et al., Horn of Africa Region: The Humanitarian Crisis and International Response 2012, US Congressional Research Service,
12 January 2012, https://sgp.fas.org/crs/row/R42046.pdf; UN OCHA, Global Humanitarian Overview 2022, https://gho.unocha.org/
195 John Bryant, Digital mapping and inclusion in humanitarian response, Humanitarian Policy Group Working Paper, October 2021,
https://cdn.odi.org/media/documents/Digital_IP_Mapping_case_study_web_EDSoP6n.pdf
• 68% of the humanitarian organisations surveyed understand the technical specifications
of satellite applications required for different use cases; however, 32% have little to zero
capacity and awareness.
The respondents to the survey identified a number of thematic domains in which they
considered the use of satellite applications to be paramount. However, there is a gap
in the level of awareness of the technical specifications which relies on professionals
having the technical expertise, knowledge/awareness of satellite data, and frequent use
of satellite information during relief efforts.
To understand how to bridge this capability gap, the survey asked respondents to assess
their perceived requirements for training on satellite data knowledge within their
organisation. The survey found:
92
Annex 3
Methodology and limitations
General methodology
For this report, the team used desk-research to conduct the landscape of satellite
applications for humanitarian emergencies. Information has been gathered from sources
including academic papers, grey literature, organisation’s websites, media publications, etc.
• Stakeholder Workshops. Three virtual workshops were held with over 40 participants
from 20 countries to gain their experiences and challenges with using satellite
technologies in humanitarian situations.
• Advisory Group. Two Advisory Group meetings were held to gain feedback on scope,
findings and conclusions. See the acknowledgments for the members of this group.
A Google Sheet database was created to capture consistent attributes for each
application. The team found and categorised over 500 satellite applications. Caribou
Space intends to make this information available publicly in 2022.
• The satellite application applies to the domains, use cases and hazards, in scope
− Private Supplier is sub-categorised into Satellite Operator & Reseller, Cloud Computing
Provider, Platform/Solution Provider, Hardware/Software Supplier, and Other.
• Domain – either Disasters, Security and Conflict, Food insecurity, Population displacement,
Health emergencies, and All (see section Domains of humanitarian emergencies)
• Use Case – up to five use cases (see section Use of satellite applications across the event
life cycle)
• Business Model
− Off-the-shelf, meaning a customer can purchase and own the satellite application from
a supplier with minimal customisation required.
− Subscription, meaning a customer pays an ongoing fee for access to the service, but
doesn’t own it.
Annex 3 Beyond Borders: Satellite Applications for Humanitarian Emergencies
− Open Access, meaning the application is available free of cost or other access barriers.
− Unknown.
• Scale
− R&D, meaning R&D and prototyping stage of application development, not yet
deployed within a user’s operational environment (ARL 1–6).
− Unknown
This data collection methodology has the following limitations and challenges:
• For some attributes, use of ‘multiple’, ‘all’, and ‘unknown’ creates issues in charting and
analytics.
Annex 3 Beyond Borders: Satellite Applications for Humanitarian Emergencies
95
Annex 4
Glossary
AOI – Area of Interest is the geographic extent GIS – A geographic information system is a system
of analysis designed to capture, store, manipulate, analyse,
manage, and present spatial or geographic data
API – Application Programming Interface
GNSS – Global Navigation Satellite Systems are
ARL – Application Readiness Level
a constellation of satellites providing signals from
AWS – Amazon Web Services space that transmit positioning and timing data to
GNSS receivers
CEOS – Committee on Earth Observation
Satellites Hazard – Circumstance, event or entity with
potential to cause harm or loss
Copernicus – The European Union’s Earth
Observation Programme, looking at our planet HDX – Humanitarian Data Exchange
and its environment for the ultimate benefit of all
HR – High resolution EO data with a resolution
European citizens
of 1m–4m
Demand-side – Community of global
Annex 4 Beyond Borders: Satellite Applications for Humanitarian Emergencies
LEO – Low Earth Orbit is an Earth-centred orbit Satellite application – Digital services and
close to the planet, often specified as an orbital products that serve a number of functions for
period of 128 minutes or less. society, the environment and the economy,
deploying satellite technology
M&E – Monitoring & Evaluation is an objective
process of understanding how a project was SatComms – Satellite communications is the
implemented, what effects it had, for whom, use of satellites to provide communication links
how and why between various points on Earth
MR – Medium resolution EO data with a Use case – A specific problem for the demand-side
resolution of 5m–25m users that is addressed by a satellite application
ODA – Official Development Assistance is a SAR – Synthetic Aperture Radar (SAR) satellites
term defined by the Development Assistance are a form of radar using the motion of the
Committee (DAC) of the Organisation for satellite to create a larger ‘virtual’ antenna that is
Economic Co-operation and Development used to create two or three-dimensional images of
(OECD) to measure aid objects, such as landscapes
PNT – Position, Navigation & Timing signals UNOSAT – The United Nations Institute for
broadcast from a global navigation satellite system Training and Research (UNITAR)’s Operational
(GNSS) from space Satellite Applications Programme
Annex 4 Beyond Borders: Satellite Applications for Humanitarian Emergencies
Resilience – The ability of a system, community VHR – Very High Resolution EO data with a
or society exposed to hazards to resist, absorb, resolution of (<1m)
accommodate, adapt to, transform and recover
VLR – Very Low Resolution EO data with a
from the effects of a hazard in a timely and
resolution of (>60m)
efficient manner, including through the
preservation and restoration of its essential basic VSAT – Very-small-aperture Terminals are
structures and functions through risk management small-sized earth stations used to transmit
and receive data, voice, and video signals over
Response – Actions taken directly before, during
a satellite communication network, excluding
or immediately after a disaster to save lives, reduce
broadcast television
health impacts, ensure public safety and meet the
basic subsistence needs of the people affected
97
© 2022 Caribou Space
www.caribou.space contact@caribou.space
Caribou
Space